Top 20 Apache Spark Interview Questions & Answers

  

Ques: 1). What is Apache Spark?

Answer:

Apache Spark is an open-source real-time processing cluster computing framework. It has a vibrant open-source community and is now the most active Apache project. Spark is a programming language that allows you to programme large clusters with implicit data parallelism and fault tolerance.

Spark is one of the Apache Software Foundation's most successful projects. Spark has unquestionably established itself as the market leader in Big Data processing. Spark is used by many enterprises on clusters with thousands of nodes. Spark is now used by large companies like as Amazon, eBay, and Yahoo!

BlockChain Interview Question and Answers

Ques: 2). What advantages does Spark have over MapReduce?

Answer:

Compared to MapReduce, Spark has the following advantages:

Spark implements processing 10 to 100 times quicker than Hadoop MapReduce due to the availability of in-memory processing, whereas MapReduce uses persistence storage for any of the data processing activities.

Unlike Hadoop, Spark has built-in libraries that allow it to do a variety of functions from the same core, including as batch processing, steaming, machine learning, and interactive SQL queries. Hadoop, on the other hand, only supports batch processing.

Hadoop is heavily reliant on discs, but Spark encourages caching and data storage in memory. Spark is capable of performing computations multiple times on the same dataset. This is called iterative computation while there is no iterative computing implemented by Hadoop.

Apache Hive Interview Questions & Answers

Ques: 3). What exactly is YARN?

Answer:

YARN is a fundamental element in Spark, similar to Hadoop, in that it provides a central and resource management platform for delivering scalable operations across the cluster. YARN, like Mesos, is a distributed container manager, but Spark is a data processing tool. Spark can be run on YARN in the same way that Hadoop Map Reduce can. Running Spark on YARN needs the use of a Spark binary distribution with YARN support.

 Apache Ambari interview Questions & Answers

Ques: 4). What's the difference between an RDD, a Dataframe, and a Dataset?

Answer:

Resilient Distributed Dataset (RDD) - RDD stands for Resilient Distributed Dataset. It is the most basic data structure in Spark, consisting of an immutable collection of records partitioned among cluster nodes. It allows us to do fault-tolerant in-memory calculations on massive clusters.

RDD, unlike DF and DS, will not keep the schema. It merely stores data. If a user wants to apply a schema to an RDD, they must first build a case class and then apply the schema to the data.

We will use RDD for the below cases:

-When our data is unstructured, A streams of text or media streams.

-When we don’t want to implement any schema.

-When we don’t care about the column name attributes while processing or accessing.

-When we want to manipulate the data with functional programming constructs than domain specific expressions.

-When we want low-level transformation, actions and control on the dataset.

DataFrame:

-Like RDD DataFrames are immutable collection of data.

-Unlike RDD DataFrame will have schema for their data making user to easily access/process large set of data which is distributed among the nodes of cluster.

-DataFrame provides a domain specific language API to manipulate distributed data and makes Spark accessible to a wider audience, beyond specialized data engineers.

-From Spark 2.x Spark DataFrames are nothing but Dataset[Row] or alias (Untyped API)

consider DataFrame as an alias for a collection of generic objects Dataset[Row], where a Row is a generic untyped JVM object

DataSet:

Dataset, by contrast, is a collection of strongly-typed JVM objects, dictated by a case class you define in Scala or a class in Java. To implement case class on RDD and use as Dataset[T].

 Apache Tapestry Interview Questions and Answers

Ques: 5). Can you explain how you can use Apache Spark along with Hadoop?

Answer:

Apache Spark provides the benefit of being Hadoop compatible. They make a powerful tech team when they work together. Using Apache Spark and Hadoop combines the processing capability of Spark with the best capabilities of HDFS and YARN. The following are some examples of how to use Hadoop Components with Apache Spark:

Batch & Real-Time Processing – MapReduce and Spark can work together, where the former handles batch processing, and the latter handles real-time processing.

HDFS – Spark can make use of the HDFS to leverage the distributed replicated storage.

MapReduce – Apache Spark can be used in conjunction with MapReduce in a similar Hadoop cluster or independently as a processing framework.

YARN – You can run Spark applications on YARN.

 Apache NiFi Interview Questions & Answers

Ques: 6). What is the meaning of the term "spark lineage"?

Answer:

• In Spark, regardless of the actual data, all dependencies between RDDs will be logged in a graph. In Spark, this is referred to as a lineage graph.

• RDD stands for Resilient Distributed Dataset, with the term "resilient" referring to fault tolerance. We can re-compute the missing or damaged partition due to node failure using RDD Lineage Graph. When we generate new RDDs based on existing RDDs, we use lineage graph spark to handle the dependencies. Each RDD keeps a pointer to one or more parents, together with metadata describing the type of relationship it has with the parent RDD.

• RDD Lineage Graph in Spark can be obtained using the ToDebugString method.

 

Ques: 7). List the various components of the Spark Ecosystem.

Answer:

These are the five types of components in the Spark Ecosystem:

GraphX: Enables graphs and graph-parallel computation.

MLib: It is used for machine learning.

Spark Core: A powerful parallel and distributed processing platform.

Spark Streaming: Handles real-time streaming data.

Spark SQL: Combines Spark's functional programming API with relational processing.

 

Ques: 8). What is RDD in Spark? Write about it and explain it.

Answer:

Resilient Distributed Dataset (RDD) is an acronym for Resilient Distributed Dataset. RDDs are a fault-tolerant core data structure in Spark that is immutable. They've disseminated partitioned datasets among the cluster nodes.

Parallelizing and referencing a data set are the two methods for constructing RDDS. Lazy evaluation is the responsibility of the RDDS. The faster processing performance in Spark is due to the lazy evaluation of RDDs.

 

Ques: 9). In Spark, how does streaming work?

Answer:

Spark gets data in real time that is separated into batches. The Spark Engine processes these batches of data, and the final stream of results is returned back in batches. DStream, or Discretized Stream, is the most basic stream unit in Spark.

 

Ques: 10). Is it feasible to access and analyse data stored in Cassandra databases using Apache Spark?

Answer: 

Yes, Apache Spark may be used to retrieve and analyse data stored in Cassandra databases. Apache Spark can access and analyse data contained in Cassandra databases using the Spark Cassandra Connector. Spark should have a functionality that allows Spark executors to communicate with local Cassandra nodes and request just local data.

Cassandra and Apache Spark can be connected to speed up queries by lowering network traffic between Spark executors and Cassandra nodes.

 

Ques: 11). What are the advantages of using Spark SQL?

Answer:

Spark SQL carries out the following tasks:

Loads data from a variety of structured datasources, such as a relational database management system (RDBMS).

It may query data using SQL commands within the Spark programme as well as JDBC/ODBC connectors from third-party tools such as Tableau.

It can also provide SQL and Python/Scala code interaction.

 

Ques: 12). What is the purpose of Spark Executor?

Answer:

The Executors are obtained on top of worker nodes in the clusters when a SparkContext is formed. Spark Executors are in charge of performing computations and storing data on the worker node. They are also in charge of returning the results to the driver.

 

Ques: 13). What are the advantages and disadvantages of Spark?

Answer:

Advantages: Spark is known for real-time data processing, which may be employed in applications such as stock market analysis, finance, and telecommunications.

Spark's stream processing allows for real-time data analysis, which can aid in fraud detection, system alarms, and other applications.

Due to its lazy evaluation mechanism and parallel processing, Spark processes data 10 to 100 times quicker.

Disadvantages: When compared to Hadoop, Spark consumes greater storage space.

The task is distributed over numerous clusters rather than taking place on a single node.

Spark's in-memory processing might be costly when dealing with large amounts of data.

When compared to Hadoop, Spark makes better use of data.

 

Ques: 14). What are some of the drawbacks of utilising Apache Spark?

Answer:

The following are some of the drawbacks of utilising Apache Spark:

There is no file management system built-in. To take benefit of a file management system, integration with other platforms such as Hadoop is essential.

Higher latency, but lower throughput as a result

It does not support the processing of real-time data streams. In Apache Spark, live data streams are partitioned into batches, which are then processed and turned back into batches. To put it another way, Spark Streaming is more like micro-batch data processing than true real-time data processing.

There are fewer algorithms available.

Record-based window requirements are not supported by Spark streaming. It is necessary to distribute work across multiple clusters instead of running everything on a single node.

Apache Spark's in-memory ability becomes a bottleneck when used for the cost-efficient processing of big data.

 

Ques: 15). Is Apache Spark compatible with Apache Mesos?

Answer:

Yes. Spark can work on Apache Mesos-managed clusters, just as it works on YARN-managed clusters. Spark may run without a resource manager in standalone mode. If it has to execute on multiple nodes, it can use YARN or Mesos.

 

Ques: 16). What are broadcast variables, and how do they work?

Answer:

Accumulators and broadcast variables are the two types of shared variables in Spark. Instead of shipping back and forth to the driver, the broadcast variables are read-only variables cached in the Executors for local referencing. A broadcast variable preserves a read-only cached version of a variable on each computer instead of delivering a copy of the variable with tasks.

Additionally, broadcast variables are utilised to distribute a copy of a big input dataset to each node. To cut transmission costs, Apache Spark distributes broadcast variables using efficient broadcast algorithms.

There is no need to replicate variables for each task when using broadcast variables. As a result, data can be processed quickly. In contrast to RDD lookup(), broadcast variables assist in storing a lookup table inside the memory, enhancing retrieval efficiency.

 

Ques: 17). In Apache Spark, how does caching work?

Answer:

Caching RDDs in Spark speeds up processing by allowing numerous accesses to the same RDD. The function of Discretized Streams, or DStreams, in Spark streaming is to allow users to cache or retain data in memory.

The functions cache () and persist(level) are used to cache data in memory and cache memory based on the storage level specified, respectively.

The persist () without the level specifier is the same as cache, which means it caches the data in memory. The persist(level) method caches data at the provided storage level, such as on disc, on the server, or in off-heap memory.

 

Ques: 18).  What exactly is Akka? What does Spark do with it?

Answer:

Akka is a Scala and Java framework for reactive, distributed, parallel, and robust concurrent applications. Akka is the foundation for Apache Spark.

When assigning tasks to worker nodes, Spark employs Akka for job scheduling and messaging between the master and the worker node.

 

Ques: 19). What applications do you utilise Spark streaming for?

Answer:

When real-time data must be streamed into the Spark programme, this method is employed. It can be broadcast from a variety of places, such as Kafka, Flume, Amazon Kinesis, and others. For processing, the streamed data is separated into batches.

Spark streaming is used to conduct real-time sentiment analysis of customers on social media sites like as Twitter and Facebook, among others.

Live streaming data processing is critical for detecting outages, detecting fraud in financial institutions, and making stock market predictions, among other things.

 

Ques: 20). What exactly do you mean when you say "lazy evaluation"?

Answer:

The way Spark works with data is intellectual. When you ask Spark to perform a task on a dataset, it follows your instructions and records them so that it doesn't forget them - but it doesn't do anything until you tell it to. When map() is invoked on an RDD, the operation is not done immediately. Transformations aren't evaluated by Spark until you use them. This aids in the overall data processing workflow optimization.

Top 20 Apache Hive Interview Questions & Answers

Ques: 1). What is Apache Hive, and how does it work?

Answer:

Apache Hive is a Hadoop-based, sophisticated warehouse project. This platform focuses on data analysis and includes data query capabilities. Hive is comparable to SQL in that it provides a user interface for querying data stored in files and database systems. And Apache Hive is a popular data analysis and querying technology used by Fortune 500 companies around the world. When it is cumbersome or inefficient to run the logic in HiveQL, Hive allows standard map reduce programmes to customise mappers and reducers (User Defined Functions UDFS).

 

BlockChain Interview Question and Answers

Ques: 2). What is the purpose of Hive?

Answer:

Hive is a Hadoop tool that allows you to organise and query data in a database-like format, as well as write SQL-like queries. It can be used to access and analyse Hadoop data using SQL syntax.

 Apache Ambari interview Questions & Answers

Ques: 3). What are the differences between local and remote meta stores?

Answer:

Local meta store: When using the Local Meta store configuration, the specified meta store service, as well as the Hive service, will run on the same Java Virtual Machine (JVM) and connect to databases that are operating in distinct JVMs, either on the same machine or on a remote machine.

Remote meta store: The Meta store service and the Apache Hive service will execute on distinct JVMs in the Remote Meta store. To connect to meta store servers, all other processes use Thrift Network APIs. You can have many meta store servers in Remote meta store for high availability.

Apache Tapestry Interview Questions and Answers

Ques: 4). Explain the core difference between the external and managed tables?

Answer:

The following are the fundamental distinctions between managed and external tables:

When a managed table is dropped, the complete metadata and table data is lost. The Hive just deletes the metadata information associated with a table and leaves the table data in HDFS, whereas the external table is quite different.

Tables that are managed and tables that are external. Hive manages the data by default when you create a table, which means it moves the data into its warehouse directory. Alternatively, you can construct an external table, which instructs Hive to refer to data stored somewhere other than the warehouse directory.

The semantics of LOAD and DROP show the difference between the two table types. Let's start with a managed table. Data loaded into a managed table is stored in Hive's warehouse directory.

 Apache NiFi Interview Questions & Answers

Ques: 5). What is the difference between a read-only schema and a write-only schema?

Answer:

A table's schema is enforced at data load time in a conventional database. The data being loaded is rejected if it does not conform to the schema. Because the data is validated against the schema when it is written into the database, this architecture is frequently referred to as schema on write.

Hive, on the other hand, verifies data when it is loaded, rather than when it is queried. This is referred to as schema on read.

Between the two approaches, there are trade-offs. Because the data does not have to be read, parsed, and serialized to disc in the database's internal format, schema on read allows for a very quick first load. A file copy or move is all that is required for the load procedure. It's also more adaptable: think of having two schemas for the same underlying data, depending on the analysis. (External tables can be used in Hive for this; see Managed Tables and External Tables.)

Because the database can index columns and compress the data, schema on write makes query time performance faster. However, it takes longer to load data into the database as a result of this trade-off. Furthermore, in many cases, the schema is unknown at load time, thus no indexes can be applied because the queries have not yet been formed. Hive really shines in these situations.

 Apache Spark Interview Questions & Answers

Ques: 6). Write a query to insert a new column? Can you add a column with a default value in Hive?

Answer:

ALTER TABLE test1 ADD COLUMNS (access_count1 int); You cannot add a column with a default value in Hive. The addition of the column has no effect on the files that support your table. Hive interprets NULL as the value for every cell in that column in order to deal with the "missing" data.

In Hive, you must effectively recreate the entire table, this time with the column filled in. It's possible that rerunning your original query with the additional column will be easier. Alternatively, you might add the column to the table you already have, then select all of its columns plus the new column's value.

Ques: 7). What is the purpose of Hive's DISTRIBUTED BY clause?

Answer:

DISTRIBUTE BY determines how map output is split between reducers. By default, MapReduce computes a hash on the keys output by mappers and uses the hash values to try to distribute the key-value pairs evenly among the available reducers. Let's say we want all of the data for each value in a column to be collected at the same time. To ensure that the records for each get to the same reducer, we can use DISTRIBUTE BY. In the same way that GROUP BY determines how reducers receive rows for processing, DISTRIBUTE BY does the same.

If the DISTRIBUTE BY and SORT BY clauses are in the same query, Hive expects the DISTRIBUTE BY clause to be before the SORT BY clause. When you have a memory-intensive job, DISTRIBUTE BY is a helpful workaround because it requires Hadoop to employ Reducers instead of having a Map-only job. Essentially, Mappers gather data depending on the DISTRIBUTE BY columns supplied, reducing the framework's overall workload, and then transmit these aggregates to Reducers.

 

Ques: 8). What occurs when you perform a query in HIVE, please?

Answer:

The Query Planner examines the query and turns it to a Hadoop Map Reduce job’s DAG (Directed Acyclic Graph).

The jobs are submitted to the Hadoop cluster in the order that the DAG suggests.

Only mappers are used for simple queries. The Input Output format is in charge of splitting an input and reading data from HDFS. After that, the data is sent to a layer called SerDe (Serializer Deserializer). The deserializer part of the SerdDe converts data as a byte stream to a structured format in this example.

Reducers will be included in Map Reduce jobs for aggregate queries. In this case, the serializer of the SerDe converts structured data to byte stream which gets handed over to the Input Output format which writes it to the HDFS.

 

Ques: 9). What is the importance of STREAM TABLE?

Answer:

When you need information from several tables, joins are useful, but when you have 1.5 billion or more data in one table and want to link it to a master table, the order of the joining tables is crucial.

Consider the following scenario: 

select foo.a,foo.b,bar.c from foo join bar on foo.a=bar.a; 

Because Hive streams the right-most table (bar) and buffers other tables (foo) in memory before executing map-side/reduce-side joins. As a result, if you buffer 1.5 billion or more records, your join query will fail since 1.5 billion records will very certainly fill up Java-Heap space exception. 

So, to overcome this limitation and free the user to remember the order of joining tables based on their record-size, Hive provides a key-word /*+ STREAMTABLE(foo) */ which tells Hive Analyzer to stream table foo.

select /*+ STREAMTABLE(foo) */ foo.a,foo.b,bar.c from foo join bar on foo.a=bar.a;

Hence, in this way user can be free of remembering the order of joining tables.

 

Ques: 10). When is it appropriate to use SORT BY instead of ORDER BY?

Answer:

When working with huge volumes of data in Apache Hive, we use SORT BY instead of ORDER BY. The fact that SORT BY comes with numerous reducers is one of the reasons for utilising it. This cuts down on the amount of time it takes to complete the task. ORDER BY, on the other hand, consists of only one reduce, which means the process takes longer than usual to complete.

 

Ques: 11). What is the purpose of Hive's Partitioning function?

Answer:

Partitioning allows users to arrange data in the Hive table in the way they want it. As a result, the system would be able to scan only the relevant data rather than the complete data set.

Consider the following scenario: Assume we have transaction log data from a business website for years such as 2018, 2019, 2020, and so on. So, in this case, you can utilise the partition key to find data for a specified year, say 2019, which will reduce data scanning by removing 2018 and 2020.

 

Ques: 12). What is dynamic partitioning and how does it work?

Answer:

The values of partition columns are exposed during runtime in dynamic partitioning, i.e. the values are known when you load data into Hive tables. The following are some examples of how dynamic partitioning is commonly used:

To move data from a non-partitioned table to a partitioned table, which reduces latency and improves sampling.

 

Ques: 13). In hive, what's the difference between dynamic and static partitioning?

Answer:

Hive partitioning is highly beneficial for pruning data during queries in order to reduce query times.

When data is inserted into a table, partitions are produced. Partitions are required depending on how data is loaded. When loading files (especially large files) into Hive tables, static partitions are usually preferred. When compared to dynamic partition, this saves time when loading data. You "statically" create a partition in the table and then move the file into that partition. 

Because the files are large, they are typically created on HDFS. Without reading the entire large file, you can retrieve the partition column value from the filename, date, and so on. In the case of dynamic partitioning, the entire large file is read, i.e. every row of data is read, and the data is partitioned into the target tables using an MR job based on specified fields in the file.

Dynamic partitions are typically handy when doing an ETL operation in your data pipeline. For example, suppose you use the transfer command to load a large file into Table X. Then you run an idle query into Table Y and split data based on table X fields such as day and country. You could wish to execute an ETL step to partition the data in Table Y's nation partition into a Table Z where the data is partitioned based on cities for a specific country alone, and so on.

Thus depending on your end table or requirements for data and in what form data is produced at source you may choose static or dynamic partition.

 

Ques: 14).What is ObjectInspector in Hive?

Answer:

The ObjectInspector is a feature that allows us to analyze individual columns and internal structure of a row object in Hive. This also provides a seamless way to access complex objects that can be stored in varied formats in the memory.

• A standard Java object
• An instance of the Java class
• A lazily initialized object

The ObjectInspector lets the users know the structure of an object and also helps in accessing the internal fields of an object.

 

Ques: 15). How does impala outperform hive in terms of query response time?

Answer:

Impala should be thought of as "SQL on HDFS," whereas Hive is more "SQL on Hadoop."

Impala, in other words, does not require Hadoop at whatsoever. It simply runs daemons on all of your nodes that store some of the data in HDFS, allowing these daemons to return data rapidly without having to conduct a full Map/Reduce process.

The rationale for this is that running a Map/Reduce operation has some overhead, so short-circuiting Map/Reduce completely can result in a significant reduction in runtime.

That stated, Impala is not a replacement for Hive; it is useful in a variety of situations. When compared to Hive, Impala does not support fault-tolerance, therefore if there is a problem during your query, it will be gone. I would recommend Hive for ETL processes where a single job failure would be costly, but Impala can be great for tiny ad-hoc queries, such as for data scientists or business analysts who just want to look at and study some data without having to develop substantial jobs.

 

Ques: 16). Explain the different components used in the Hive Query processor?

Answer:

Below mentioned is the list of Hive Query processors:

• Metadata Layer (ql/metadata)
• Parse and Semantic Analysis (ql/parse)
• Map/Reduce Execution Engine (ql/exec)
• Sessions (ql/session)
• Type Interfaces (ql/typeinfo)
• Tools (ql/tools)
• Hive Function Framework (ql/udf)
• Plan Components (ql/plan)
• Optimizer (ql/optimizer)

 

Ques: 17). What is the difference between Hadoop Buffering and Hadoop Streaming?

Answer:

Using custom made python or shell scripts to implement your map-reduce logic is known as Hadoop Streaming. (Use the Hive TRANSFORM keyword, for example.)

In this context, Hadoop buffering refers to the phase in a map-reduce job of a Hive query with a join when records are read into the reducers after being sorted and grouped by the mappers. The author explains why you should order the join clauses in a Hive query so that the largest tables come last; this helps Hive implement joins more efficiently.

 

Ques: 18). How will the work be optimised by the map-side join?

Answer:

Let's pretend we have two tables, one of which is a little table. A Map Reduce local job will be generated before the original join Map Reduce task, which will read data from HDFS and put it into an in-memory hash table. It serialises the in-memory hash table into a hash table file after reading it.

The data in the hash table file is then moved to the Hadoop distributed cache, which populates these files to each mapper's local disc in the following stage, while the original join Map Reduce process is running. As a result, all mappers can reload this permanent hash table file into memory and perform the join operations as previously. 

The optimised map join's execution sequence is depicted in the diagram below. The short table just has to be read once after optimization. In addition, if many mappers are operating on the same system, the distributed cache only needs to send a single copy of the hash table file to this machine.

Advantages of using Map-side join:

Using Map-side join reduces the cost of sorting and combining data in theshuffle and reduces stages. The map-side join also aids task performance by reducing the time it takes to complete the assignment.

Disadvantages of Map-side join:

It is only suitable for use when one of the tables on which the map-side join operation is performed is small enough to fit into memory. As a result, performing a map-side join on tables with a lot of data in each of them isn't a good idea.

 

Ques: 19).What type of user defined functions exists in HIVE?

Answer:

A UDF operates on a single row and produces a single row as its output. Most functions, such as mathematical functions and string functions, are of this type.

A UDF must satisfy the following two properties:

• A UDF must be a subclass of org.apache.hadoop.hive.ql.exec.UDF.
• A UDF must implement at least one evaluate() method.

 

A UDAF works on multiple input rows and creates a single output row. Aggregate functions include such functions as COUNT and MAX.

• A UDAF must satisfy the following two properties:
• A UDAF must be a subclass of org.apache.hadoop.hive.ql.exec.UDAF;
• An evaluator must implement five methods:
• init()
• iterate()
• terminatePartial()
• merge()
• terminate()

• A UDTF operates on a single row and produces multiple rows — a table — as output.
• A UDTF must be a subclass of org.apache.hadoop.hive.ql.udf.generic.GenericUDTF;
• A custom UDTF can be created by extending the GenericUDTF abstract class and then implementing the initialize, process, and possibly close methods.
• The initialize method is called by Hive to notify the UDTF the argument types to expect.
• The UDTF must then return an object inspector corresponding to the row objects that the UDTF will generate.
• Once initialize() has been called, Hive will give rows to the UDTF using the process() method.
• While in process(), the UDTF can produce and forward rows to other operators by calling forward().
• Lastly, Hive will call the close() method when all the rows have passed to the UDTF.

 

Ques: 20). Is the HIVE LIMIT clause truly random?

Answer:

Although the manual claims that it returns rows at random, this is not the case. Without any where/order by clause, it returns "selected rows at random" as they occur in the database. This doesn't imply it's truly random (or randomly picked), but it does suggest that the order in which the rows are returned can't be predicted.

It returns the last 5 rows of whatever you're picking from as soon as you slap an order by x DESC limit 5 on there. You'd have to use something like order by rand() LIMIT 1 to get rows returned at random.

However, if your indexes aren't set up correctly, it can slow things down. I usually do a min/max to get the IDs on the table, then a random number between them, then choose those records (in your instance, just one), which is usually faster than letting the database do the work, especially on a huge dataset.

Top 20 Apache Ambari interview Questions & Answers

  

Ques: 1). Describe Apache Ambari's main characteristics.

Answer:

Apache Ambari is an Apache product that was created with the goal of making Hadoop applications easier to manage. Ambari assists in the management of the Hadoop project.

• Provisioning is simple.
• Project management made simple
• Monitoring of Hadoop clusters
• Availability of a user-friendly interface
• Hadoop management web UI
• RESTful API support

 Apache Tapestry Interview Questions and Answers

Ques: 2). Why do you believe Apache Ambari has a bright future?

Answer:

With the growing need for big data technologies like Hadoop, we've witnessed a surge in data analysis, resulting in gigantic clusters. Companies are turning to technologies like Apache Ambari for better cluster management, increased operational efficiency, and increased visibility. Furthermore, we've noted how HortonWorks, a technology titan, is working on Ambari to make it more scalable. As a result, learning Hadoop as well as technologies like Apache Ambari is advantageous.

 Apache NiFi Interview Questions & Answers

Ques: 3). What are the core benefits for Hadoop users by using Apache Ambari?

Answer: 

The Apache Ambari is a great gift for individuals who use Hadoop in their day to day work life. With the use of Ambari, Hadoop users will get the core benefits:

1. The installation process is simplified
2. Configuration and overall management is simplified
3. It has a centralized security setup process
4. It gives out full visibility in terms of Cluster health
5. It is extensively extendable and has an option to customize if needed.

 Apache Spark Interview Questions & Answers

Ques: 4). What Are The Checks That Should Be Done Before Deploying A Hadoop Instance?

Answer:

Before actually deploying the Hadoop instance, the following checklist should be completed:

• Check for existing installations
• Set up passwordless SSH
• Enable NTP on the clusters
• Check for DNS
• Disable the SELinux
• Disable iptables

 Apache Hive Interview Questions & Answers

Ques: 5 As a Hadoop user or system administrator, why should you choose Apache Ambari?

Answer:

Using Apache Ambari can provide a Hadoop user with a number of advantages.

A system administrator can use Ambari to – Install Hadoop across any number of hosts using a step-by-step guide supplied by Ambari, while Ambari handles Hadoop installation setup.

Using Ambari, centrally administer Hadoop services across the cluster.

Using the Ambari metrics system, efficiently monitor the state and health of a Hadoop cluster. Furthermore, the Ambari alert framework sends out timely notifications for any system difficulties, like as disc space issues or node status.

 

Ques: 6). Can you explain Apache Ambari architecture?

Answer:

Apache Ambari consists of following major components-

• Ambari Server
• Ambari Agent
• Ambari Web

Apache Ambari Architecture

The all metadata is handled by the Ambari server, which is made up of a Postgres database instance as indicated in the diagram. The Ambari agent is installed on each computer in the cluster, and the Ambari server manages each host through it.

An Ambari agent is a member of the host that delivers heartbeats from the nodes to the Ambari server, as well as numerous operational metrics, to determine the nodes' health condition.

Ambari Web UI is a client-side JavaScript application that performs cluster operations by regularly accessing the Ambari RESTful API. Furthermore, using the RESTful API, it facilitates asynchronous communication between the application and the server.

 

Ques: 7). Apache Ambari supports how many layers of Hadoop components, and what are they?

Answer: 

Apache Ambari supports three tiers of Hadoop components, which are as follows:

1. Hadoop core components

• Hadoop Distributed File System (HDFS)
• MapReduce

2. Essential Hadoop components

• Apache Pig
• Apache Hive
• Apache HCatalog
• WebHCat
• Apache HBase
• Apache ZooKeeper

3. Components of Hadoop support

• Apache Oozie
• Apache Sqoop
• Ganglia
• Nagios

 

Ques: 8). What different sorts of Ambari repositories are there?

Answer: 

Ambari Repositories are divided into four categories, as below:

1. Ambari: Ambari server, monitoring software packages, and Ambari agent are all stored in this repository.
2. HDP-UTILS: The Ambari and HDP utility packages are stored in this repository.
3. HDP: Hadoop Stack packages are stored in this repository.
4. EPEL (Enterprise Linux Extra Packages): The Enterprise Linux repository now includes an extra set of software.

 

Ques: 9). How can I manually set up a local repository?

Answer:

When there is no active internet connection available, this technique is used. Please follow the instructions below to create a local repository:

1. First and foremost, create an Apache httpd host.
2. Download a Tarball copy of each repository's entire contents.
3. After it has been downloaded, the contents must be extracted.

 

Ques: 10). What is a local repository, and when are you going to utilise one?

Answer:

A local repository is a hosted place for Ambari software packages in the local environment. When the enterprise clusters have no or limited outbound Internet access, this is the method of choice.

 

Ques: 11). What are the benefits of setting up a local repository?

Answer: 

First and foremost by setting up a local repository, you can access Ambari software packages without internet access. Along with that, you can achieve benefits like –

Enhanced governance with better installation performance

Routine post-installation cluster operations like service start and restart operations

 

Ques: 12). What are the new additions in Ambari 2.6 versions?

Answer:

Ambari 2.6.2 added the following features:

• It will protect Zeppelin Notebook SSL credentials
• We can set appropriate HTTP headers to use Cloud Object Stores with HDP
• Ambari 2.6.1 added the following feature:
• Conditional Installation of  LZO packages through Ambari
• Ambari 2.6.0 added the following features:
• Distributed mode of Ambari Metrics System’s (AMS) along with multiple Collectors
• Host Recovery improvements for the restart
• moving masters with minimum impact and scale testing
• Improvement in Data Archival & Purging in Ambari Infra

 

Ques: 13). List Out The Commands That Are Used To Start, Check The Progress And Stop The Ambari Server?

Answer :

The following are the commands that are used to do the following activities:

To start the Ambari server

ambari-server start

To check the Ambari server processes

ps -ef | grep Ambari

To stop the Ambari server

ambari-server stop

 

Ques: 14). What all tasks you can perform for managing host using Ambari host tab?

Answer: 

Using Hosts tab, we can perform the following tasks:

• Analysing Host Status
• Searching the Hosts Page
• Performing Host related Actions
• Managing Host Components
• Decommissioning a Master node or Slave node
• Deleting a Component
• Setting up Maintenance Mode
• Adding or removing Hosts to a Cluster
• Establishing Rack Awareness

 

Ques: 15). What all tasks you can perform for managing services using Ambari service tab?

Answer: 

Using Services tab, we can perform the following tasks:

• Start and Stop of All Services
• Display of Service Operating Summary
• Adding a Service
• Configuration Settings change
• Performing Service Actions
• Rolling Restarts
• Background Operations monitoring
• Service removal
• Auditing operations
• Using Quick Links
• YARN Capacity Scheduler refresh
• HDFS management
• Atlas management in a Storm Environment

 

Ques: 16). Is there a relationship between the amount of free RAM and disc space required and the number of HDP cluster nodes?

Answer: 

Without a doubt, it has. The amount of RAM and disc required depends on the number of nodes in your cluster. In typically, 1 GB of memory and 10 GB of disc space are required for each node. Similarly, for a 100-node cluster, 4GB of memory and 100GB of disc space are required. To get all of the details, you'll need to look at a specific version.

 

Ques: 17). What tasks you can skill for managing services using the Ambari subsidiary bank account?

Answer: 

using the Services report, we can do the bearing in mind tasks:

• Start and Stop of All Services
• Display of Service Operating Summary
• Adding a Service
• Configuration Settings regulate
• Performing Service Actions
• Rolling Restarts
• Background Operations monitoring
• Service removal
• Auditing operations
• Using Quick Links
• YARN Capacity Scheduler refresh
• HDFS presidency
• Atlas approach in a Storm Environment

 

Ques: 18). What is the best method for installing the Ambari agent on all 1000 hosts in the HDP cluster?

Answer: 

Because the cluster contains 1000 nodes, we should not manually install the Ambari agent on each node. Instead, we should set up a password-less ssh connection between the Ambari host and all of the cluster's nodes. To remotely access and install the Ambari Agent, Ambari Server hosts employ SSH public key authentication.

 

Ques: 19). What can I do if I have admin capabilities in Ambari?

Answer: 

Becoming a Hadoop Administrator is a difficult job. On HadoopExam.com, you can find all of the available Hadoop Admin training for HDP, Cloudera, and other platforms (visit now). You can create a cluster, manage the users in that cluster, and create groups if you are an Ambari Admin. All of these permissions are granted to the default admin user. You can grant the same or different permissions to another user even if you are an Amabari administrator.

 

Ques: 20).  How is recovery achieved in Ambari?

Answer:

Recovery happens in Ambari in the moreover ways:

Based in remarks to activities

In Ambari after a restart master checks for pending undertakings and reschedules them previously all assimilation out is persisted here. Also, the master rebuilds the come clean machines at the back there is a restart, as the cluster market is persisted in the database. While lawsuit beautifies master actually catastrophe in the in front recording their take keep busy, along amid there is a race condition. The events, on the other hand, should be idempotent, which is a unique consideration. And the master restarts any behavior that has not been marked as occurring or has failed in the database. These persistent behaviors are seen in Redo Logs.

Based approaching the desired make known

While the master attempts to make the cluster flesh and blood publicise, you will be encircled by more to in as per the intended freshen appendix, as the master persists in the desired own going in savings account to for of the cluster.

Top 20 Edge Computing Interview Questions & Answers

Ques: 1). What is edge computing, and how does it work?

Answer:

With the passage of time, technology tends to become smaller and faster. As a result, previously "dumb" items such as light bulbs and door locks can now contain modest CPUs and RAM. They can perform calculations and provide information on usage. This computing enables analytics to be performed at the network's most granular levels, often known as the edge.

Edge computing puts processing power closer to the end user or the data source. In practise, this implies relocating computation and storage from the cloud to a local location, such as an edge server. Read more about edge computing in our overview.

 

Ques: 2). Is the footprint of this edge service appropriate for my needs?

Answer:

Different edge computing applications may have drastically different needs for geographic coverage and proximity. Consider the requirements of your project. Edge computer nodes could be located within or near each factory, but only for a limited number of locations.

The creator of an augmented reality programme that customers can use in stores to get real-time product ratings and pricing comparisons might want edge nodes on every street corner, or as near to that as possible.

 

Ques: 3). Why Edge Computing?

Answer:

This technique optimises bandwidth efficiency by analysing data at the edge, as opposed to the cloud, which requires data transfer from the IoT, which requires high bandwidth, making it beneficial for application in remote locations at low cost. It enables smart applications and devices to react to data practically simultaneously, which is critical in business and self-driving automobiles. It can process data without putting it on a public cloud, which assures complete security.

While on an extended network, data may become corrupt, compromising the data's dependability for companies to use. The utilisation of cloud computing is limited by data computation at the edge.

 

Ques: 4).  What are the main Key Benefits and services Of Edge Computing?

Answer:

• Faster response time.
• Security and Compliance.
• Cost-effective Solution.
• Reliable Operation With Intermittent Connectivity.

Edge Cloud Computing Services:

• IOT (Internet Of Things)
• Gaming
• Health Care
• Smart City
• Intelligent Transportation
• Enterprise Security

 

Ques: 5). Is there really a need for that much computation at the edge?

Answer:

Another way to phrase this question is: Which data-intensive tasks would benefit the most from network offloading? Not all applications will be eligible, and many will require data aggregation that is beyond the capability of local computing. Look for situations where processing data closer to the consumer or data source would be more efficient. These three, according to Steven Carlini, are the best prospects for edge computing.

 

Ques: 6). Is there really a need for that much computation at the edge?

Answer:

Another way to phrase this question is: Which data-intensive tasks would benefit the most from network offloading? Not all applications will be eligible, and many will require data aggregation that is beyond the capability of local computing. Look for situations where processing data closer to the consumer or data source would be more efficient. These three, according to Steven Carlini, are the best prospects for edge computing.

 

Ques: 7). How much storage should be available at the edge?

Answer:

Large volumes of data that would have been saved in the cloud will now be stored locally thanks to edge computing. While storage technology is inexpensive, management costs are not. Will the cost of keeping and managing device data at the edge justify the move? How will edge devices be protected?

Processing data at the edge, rather than uploading raw data to the cloud, may be a better way to secure user privacy. Edge computing's dispersed nature, on the other hand, renders intelligent edge devices more susceptible to malware outbreaks and security breaches.

 

Ques: 8). Why is it important to concentrate on edge computing right now?

Answer:

Edge is ripe now, thanks to new technology and demand for new applications. Consumers seek reduced latency for content-driven experiences, while businesses need local processing for security and redundancy. If you're interested in learning more about where edge computing is going, check out our article on the future of edge computing.

 

Ques: 9). What kind of apps, services, or business strategies would your edge computing platform deliver?

Answer:

Determine which workloads should run on the edge rather than in a central location, if you haven't previously, says Yugal Joshi, vice president of Everest Group. IT leaders should also look into whether any existing initiatives (such as IoT or AI) could benefit from edge processing.

 

Ques: 10). Will the organization's operating model have to alter as a result of edge computing?

Answer:

The usage of edge computing to support operational technologies is common. In such circumstances, technology leaders must determine who will own and manage the edge environment, whether greater alignment between the operating and information technology groups is required, and how performance will be monitored.

 

Ques: 11). What is the distinction between edge computing, cloud computing, and fog computing?

Answer:

Data collection, storage, and calculation are all done on edge devices in edge computing.

Cloud computing is the storing and computation of data on servers that are primarily more powerful and connected to edge devices. The edge devices transfer their data across the network to the cloud, where it is processed by a more sophisticated system.

Fog computing is a hybrid of the two approaches. The cloud servers are sometimes too far away from the edge devices for data analytics to happen quickly enough. As a result, a fog computing intermediary device is set up as a hub between two fog computing devices. This device does the computation and analytics required by the edge device.

 

Ques: 12). What role does a database play in edge computing?

Answer:

A device on the edge must be able to store and manage the data it generates efficiently. These devices have very little CPU and storage space, and they may power cycle frequently and unpredictably. A database system is the only means to store and use data in a secure manner. Additionally, the data may need to be easily transferred to a cloud system or accessed from a remote location. A database system  with SymmetricDS can provide a developer with a simple set of APIs to accomplish this.

 

Ques: 13). What is the sturdiness of this edge solution? How will the edge provider ensure that the application recovers if it fails?

Answer:

As businesses move beyond experimenting with edge computing to leveraging it for more significant applications, questions like these will become increasingly essential. To mitigate the risks of the innovative edge components, IT architects will want to use tried-and-true technologies whenever possible. Service level agreements and quality of service guarantees are important to business leaders. Even so, there will be setbacks.

 

Ques: 14). What is our long-term plan for managing edge resources?

Answer:

It's difficult enough to manage network and computer resources that are split between company data centres and the cloud. The difficulty could be amplified with edge computing.

You should inquire about what systems management resources an edge service provider provides, as well as how well-known systems management software vendors are addressing the unique aspects of edge computing.

There's also the issue of labour division: how much control will the enterprise have over how software is deployed to and updated on edge nodes? How much of that will it entrust to a third-party service provider?Will the enterprise even have the option of exercising control over the management of cloud nodes, or will the service provider consider that its own business?

 

Ques: 15). What safeguards do we have in place to avoid becoming enslaved to this cutting-edge solution?

Answer:

For the most part, open source software and open standards have prevailed in the cloud, and they're likely to win on the edge for the same reasons, according to Drobot. Open internet technologies are the most adaptable and portable, making them popular among clients as well as cloud providers who need to improve their solutions on a regular basis. He predicts that the same dynamics will apply to edge computing. The biggest exceptions so far have been related to edge computing resource metering and billing technology. Technology for managing edge computing that is specific to a particular vendor’s environment could make it harder to move your applications elsewhere.

 

Ques: 16). How might edge computing aid in the real-time visualisation of my business?

Answer:

Because data is handled in parallel across several edge nodes, edge computing allows industrial data to be processed more efficiently. Furthermore, because data is computed at the edge, delay from a round trip across the local network, to the cloud, and back is not required. Edge computing is hence well suited for real-time applications. Edge computing can assist in the prevention of equipment failure by detecting and forecasting when faults may occur, allowing operators to respond earlier. Real-time KPIs can provide decision makers with a complete picture of their system's state. Identifying which information is most valuable to receive in real-time can scope edge computing projects to focus on what’s important.

 

Ques: 17). How can I put Machine Learning to work at the edge?

Answer:

At the edge, machine learning algorithms can reduce raw sensor data by removing duplicates and other noise. Machine Learning can greatly reduce the amount of data that has to be transferred over local networks or kept in the cloud or other database systems by identifying useful information and discarding the rest. Machine learning in edge installations ensures cheaper running costs and more efficient operation of downstream applications.

 

Ques: 18). Where do you see possibilities for integrating with existing systems?

Answer:

According to a survey conducted by IDC Research, 60% of IT workers have five or more analytical databases, and 25% have more than ten. Edge computing allows these external systems to be integrated into a single real-time experience. Edge computing systems can easily consider other systems as new nodes in the system by employing bridges and connections, whereas integration has previously been a big difficulty. As a result, seeing integration opportunities early on can help you get the most out of your edge computing solution.

 

Ques: 19). What kinds of costly incidents may be avoided if I was alerted sooner?

Answer:

Edge computing architectures' real-time advantages can help minimise costly downtime and other unintended consequences. You may more effectively prioritise the desired objectives for your edge computing project by analysing which events can be the most disruptive to your organisation. Edge computing can assist identify the conditions that cause failure in real-time and enable operators to intervene sooner, whether your objective is to reduce downtime, develop an effective predictive maintenance strategy, or ensure that logistical operations are made more efficient.

 

Ques: 20). What can I do to make it more secure?

Answer:

Edge deployments are complicated, as each node adds to the vulnerability surface area. As a result, security planning is vital to the success of any edge computing project. Edge computing enables the encryption of critical data at the point of origin, ensuring an end-to-end security solution. Additional security steps can be taken by separating edge services from the rest of the programme, guaranteeing that even if one node is hacked, the remainder of the application can continue to function normally.

Top 20 Apache ActiveMQ Interview Questions & Answers

 

Ques: 1). What exactly is ActiveMQ?

Answer: 

Apache Message-oriented middleware (MOM) is a type of software that transmits messages between applications, and ActiveMQ is one of them. ActiveMQ facilitates loose coupling of elements in an IT system using standards-based, asynchronous communication, which is frequently basic to business messaging and distributed applications. Messages are translated from sender to receiver using ActiveMQ. Instead of requiring both the client and the server to be online at the same time in order to interact, it can connect numerous clients and servers and allow messages to be queued.

AWS Lambda Interview Questions & Answers

Ques: 2). In Apache ActiveMQ, what are clusters?

Answer: 

Load balancing of messages on a queue between consumers is supported by ActiveMQ in a stable and high-performance manner. This scenario is known as the competing consumers pattern in corporate integration. The principle is illustrated in the diagram below: Interview questions for Activemq

The burden is distributed in an extremely fluid manner. In high-load periods, more consumers might be provisioned and joined to the queue without changing any queue setup, as the new consumer would behave like any other competing consumer. Better availability than load-balanced systems. To determine whether real-servers are offline, load balancers often use a monitoring system. A failed consumer will not compete for messages if there are competing consumers, so messages will not be given to it even if it is not monitored.

AWS Cloudwatch interview Questions & Answers

Ques: 3). What Is The Difference Between ActiveMQ And AmQP?

Answer: 

The Advanced Message Queue Protocol is a wire-level protocol for client-to-messaging-broker communication that serves as a specification for how messaging clients and brokers will interact.

AMQP is a message protocol rather than a messaging system like ActiveMQ.

Open wire protocols, such as OpenWire, a fast binary format, are supported by AMQP.

Stomp is a text-based protocol that is simple to implement.

MQTT is a little binary format designed for restricted devices over a shaky network.

AWS Cloud Support Engineer Interview Question & Answers

Ques: 4). What distinguishes ActiveMQ from other messaging systems?

Answer: 

• It is a Java messaging service implementation, therefore it contains all of Java's features.
• Extremely persistent
• It has a high level of security and authentication.
• Various brokers can form a cluster and collaborate with one another.
• ActiveMQ offers a number of client APIs in a range of languages.

AWS Solution Architect Interview Questions & Answers

Ques: 5). What are the most important advantages of ActiveMQ?

Answer: 

• Allows users to combine many languages with various operating systems.
• Allows for location transparency.
• Communication that is both reliable and effective
• It's simple to scale up and offers asynchronous communication.
• Reduced coupling

AWS RedShift Interview Questions & Answers

Ques: 6). What are ActiveMQ's biggest drawbacks?

Answer: 

It's a complicated mechanism that only allows one thread per connection.

AWS DevOps Cloud Interview Questions & Answers

Ques: 7), In ActiveMQ, what is a topic?

Answer: 

Virtual Topics are a hybrid of Topics and Queues, with listeners consuming messages from the queues as messages to the topic.

ActiveMQ assists in replicating and duplicating every message from the topic to the actual consumers queues.

AWS(Amazon Web Services) Interview Questions & Answers

Ques: 8). What is the difference between Activemq and Fuse Message Broker?

Answer: 

Multiple Protocol Messaging is a Java-based message broker that supports industry-standard protocols and allows users to choose from a wide range of client languages, including JavaScript, C, C++, and Python.

Fuse Message Broker is a distributor of FuseSource's Apache ActiveMQ, which it develops and updates as part of the Apache ActiveMQ community.

Bug fixes are more likely to come from the Fuse Broker release than from an official Apache ActiveMQ release.

AWS Database Interview Questions & Answers

Ques: 9). What exactly is KahaDB?

Answer: 

KahaDB is a file-based persistence database that runs on the same machine as the message broker. It has been designed to be persistent in a short amount of time. Since ActiveMQ 5.4, it has been the default storage mechanism. Compared to its predecessor, the AMQ Message Store, KahaDB consumes fewer file descriptors and recovers faster.

 

Ques: 10). What exactly is LevelDB?

Answer: 

LevelDB is a somewhat faster index than KahaDB, with slightly better performance figures. The LevelDB store will allow replication in forthcoming ActiveMQ releases.

 

Ques: 11). What's the difference between RabbitMQ and ActiveMQ?

Answer: 

ActiveMQ is a Java-scripted open-source message broker that is built on the Java Message Service client. The RabbitMQ protocol is based on the Advanced Message Queuing protocol.

 

Ques: 12). What are the benefits of using a combination of topics and queues instead of traditional topics?

Answer: 

There will be no lost communications even if a customer is offline. All messages are copied to the queues that have been registered by ActiveMQ.

A dead letter queue will be set up if a customer is unable to process a message. Without affecting the other consumers, the consumer can be resolved and the message forwarded to his own dedicated queue.

To implement a load balancing mechanism we can register multiple instances of a consumer on a queue.

 

Ques: 13). If the ActiveMQ server is unavailable, what should I do?

Answer: 

This begins with ActiveMQ's storage mechanism. Non-persistent messages are stored in memory under typical conditions, while persistent messages are stored in files, with their maximum limitations set in the configuration file's node. When the number of non-persistent messages reaches a specific threshold and memory becomes scarce, ActiveMQ will write the non-persistent messages in memory to a temporary file to free up space Despite the fact that they are all saved in files, the distinction between persistent messages and non-persistent temporary files is that persistent messages will be restored from the file after restart, whereas non-persistent temporary files would be removed immediately.

 

Ques: 14). What happens if the file size exceeds the configuration's maximum limit?

Answer: 

Set a 2GB persistent file limit and mass-produce persistent messages until the file exceeds its limit. The producer is currently prohibited, but the consumer can connect and consume the message as usual. The producer can continue to transmit messages after a portion of the message has been eaten and the file has been erased to make room, and the service will automatically revert to normal.

Set a 2GB limit on temporary files, mass-produce non-persistent messages, and write temporary files. When the maximum limit is reached, the producer is blocked, and consumers can still connect but not consume messages, or consumers who were previously sluggish consumers suddenly consume Stop. The complete system is linked, but it is unable to give services, causing it to hang.

 

Ques: 15). What is message-oriented middleware, and how does it work?

Answer: 

Message-oriented middleware (MOM) is a software or hardware framework that allows distributed systems to send and receive messages. MOM simplifies the development of applications that span different operating systems and network protocols by allowing application modules to be distributed across heterogeneous platforms. The middleware establishes a distributed communications layer that hides the intricacies of the multiple operating systems and network interfaces from the application developer.

 

Ques: 16). What is the benefit of Activemq over other options such as databases?

Answer: 

Activemq is a messaging system that allows two distributed processes to communicate successfully. It can keep messages in a database to communicate between processes, but you'd have to erase them as soon as they were received. For each message, this means a row insert and remove. When you try to scale that up to hundreds of messages per second, databases start to break down.

Message-oriented middleware, such as ActiveMQ, is designed to handle these scenarios. They assume that messages will be erased promptly in a healthy system and can make optimizations to prevent the overhead. It can also push messages to consumers rather than requiring them to poll for fresh messages via SQL queries. This minimises the amount of time it takes for new messages to be processed into the system.

 

Ques: 17). What are some of the platforms supported by ActiveMQ?

Answer: 

Some of the common platforms supported by ActiveMQ include:

Any java platform that has an update of 5.0 or more.

J2EE 1.4 is another platform

JMS 1.1

JCA 1.5 resource adaptor

 

Ques: 18). Make a distinction between ActiveMQ and Mule.

Answer: 

ActiveMQ is a messaging service with a lot of options for both the broker and the client. Mule, on the other side, is an ESB that may provide executive functionality to merely the broker by exchanging messages between various software components.

Mule's architecture is such that it is designed to provide a programming configuration that is feasible for integrating applications between a database and an operating system. Mule, on the other hand, does not support any form of native messaging system, hence it is typically used in conjunction with ActiveMQ. the user is required to introduce different and unique frameworks to define various boundaries for connectivity.

 

Ques: 19). What is the process for dealing with an application server using JMS connections?

Answer: 

The server session is created with the help of an application server, which then stores them in a pool. An association buyer uses the server's session to place messages in JMS sessions. The JMS session is created by a server session. The messaging audience is created by an application produced by application software engineers.

 

Ques: 20). What distinguishes ActiveMQ from the spread toolkit?

Answer: 

Spread Toolkit is a C++ library for informing, with only rudimentary support for JMS. It does not support robust informing, exchanges, XA, or JMS 1.1 in its entirety. It's also depending on a locally installed version of Spread inspiration. Apache ActiveMQ, on the other hand, is the JMS provider used in Apache Geronimo. It is J2EE 1.4 certified in Geronimo and is a completely pure version of the Java programming language. ActiveMQ supports real-time and persistent messaging, exchanges, XA, J2EE 1.4, JMS 1.1, JCA 1.5, and a slew of other features like Message Groups and Clustering.