# Apache Kudu Reviews **Vendor:** The Apache Software Foundation **Category:** [Columnar Databases](https://www.g2.com/categories/columnar-databases) **Average Rating:** 4.1/5.0 **Total Reviews:** 13 ## About Apache Kudu Apache Kudu is a free and open source column-oriented data store of the Apache Hadoop ecosystem. ## Apache Kudu Reviews ### 1. Tryst With Kudu in Production Deployment for a Telco **Rating:** 4.0/5.0 stars **Reviewed by:** Satheesh V. | Senior Solutions Architect, Mid-Market (51-1000 emp.) **Reviewed Date:** December 06, 2020 **What do you like best about Apache Kudu?** 1. Implementing Lambda Architecture (Both Batch & Real Time Streaming) with Kudu is quite straightforward. We additionally used Streamsets as the Ingestion Platform which has quite a good coupling with Kudu. 2. Makes Real Time Analytics Quite Starightforward. We used Kudu to run multiple Real Time Campaigns. 3. Tailor Made for Implementing Data Warehouses within a Big Data Environment. 4. Nice Upsert Functionality. **What do you dislike about Apache Kudu?** 1. Partition Limitation - Is limited to 2000 tablets/tablet server. 2. Faces Random Timeouts while approaching the tablet number limitation (max 2000/server). 3. Needs a Conformed Schema. No automatic handling of drifting schemas. 4. Needs a Primary Key for every table. 5. #3 & #4 are not really cons per se. **Recommendations to others considering Apache Kudu:** Use it if the Datawarehouse is a strong use case within Big Data enviroments & try to be within the partition/tablet limitation and you should be good to go. **What problems is Apache Kudu solving and how is that benefiting you?** 1. Real Time Campaigns. 2. Real Time Lookups & Transaction Enrichment. 3. Data Warehouse Implementation is quite straight forward with Apache Kudu. ### 2. A good choice for efficient on-the-fly big data analytics + write random access **Rating:** 4.0/5.0 stars **Reviewed by:** Araceli M. | Data Engineer, Enterprise (> 1000 emp.) **Reviewed Date:** December 15, 2020 **What do you like best about Apache Kudu?** To my mind, what I found most helpful about Apache Kudu is the possibility to make quite complex data aggregations through SQL engines while also being able to efficiently perform updates **What do you dislike about Apache Kudu?** Probably two downsides of Apache Kudu to take into account are partitioning configuration, as it's quite hard to find the best partitioning set up for the particular use case performance, and also the fact that Kudu does not balance charge if a new node is added to the cluster. **Recommendations to others considering Apache Kudu:** Be ready for expend quite a few hours fine-tunning Apache Kudu partitioning and figuring out the best replication factor for your particular use case. **What problems is Apache Kudu solving and how is that benefiting you?** While working as a Financial Data Engineer at Santander CIB we leveraged Apache Kudu potential for efficient online data analytics for financial derivatives calculation in an on-premise cluster. We realised Kudu was suitable for this kind of computation engine that requires both high throughput and low latency because of its characteristics. It was also a good choice since it was a Cloudera cluster and has a really good integration capability with HDFS and Impala. ### 3. I am happy with Kudu except for few exceptions. **Rating:** 4.5/5.0 stars **Reviewed by:** Vikash S. | Sr AIML & Data Engineer, Enterprise (> 1000 emp.) **Reviewed Date:** December 01, 2020 **What do you like best about Apache Kudu?** Its quite good while reading and loading and best columnar database on top of hadoop. This kudu come with Cloudera so all component in one place. integration with impala script and compatible with parquet file format make it make more dynamic. Its dynamic partition also help during debugging. Kudu hive metastore integration is also make hive and kudu table accessible from same platform. The incremental data is replicated without any issue for end users. It support almost all ANSI sql so look very familiar. **What do you dislike about Apache Kudu?** Kudu partitioning is done only on primary key which is not always suitable. Some cases we have time stamp as partition key but could not be primary key so reading huge table is concern here. So this need to be change in future . There is concern over Kudu Context concern with impala integration. Currently JDBC connectivity has still some issue for huge data. Its camecase table name issue.Kudu tables with a column name containing upper case or non-ascii characters may not be used as an external table in Impala. Columns may be renamed in Kudu to work around this issue. **Recommendations to others considering Apache Kudu:** Go for kudu if you have Hadoop based ecosystem and loading and reading the data. **What problems is Apache Kudu solving and how is that benefiting you?** We have creating cube for our dashboard. This kudu come with Cloudera so all component in one place. integration with impala script and compatible with parquet file format make it make more dynamic. Kudu hive metastore integration is also make hive and kudu table accessible from same platform. ### 4. Apache Kudu vs Hive, speed of execution. **Rating:** 5.0/5.0 stars **Reviewed by:** Bharat M. | Specialist Architect , Enterprise (> 1000 emp.) **Reviewed Date:** December 03, 2020 **What do you like best about Apache Kudu?** Speed of execution, specially with comparison to Apache Hive. Flexibility to work as a regular RDBMS table. It can be used for huge datamart with CDC logic at ease. **What do you dislike about Apache Kudu?** You have to decide the primary key for each and every objects. So, more focus on the Data Modelling aspect. **What problems is Apache Kudu solving and how is that benefiting you?** We were trying to create DataMart for a huge financial data and where we were looking for a CDC as well as flexibility to update the data as per the business need. we tried Hive but it didn't work as expected. Then we tried Kudu and its performing well. ### 5. Working with Apache kudu **Rating:** 2.5/5.0 stars **Reviewed by:** Aman K. | SDET, Mid-Market (51-1000 emp.) **Reviewed Date:** December 05, 2020 **What do you like best about Apache Kudu?** Easily accessible from APACHE Impala query engine ,So building dashboard and OLAP experience is seamless. **What do you dislike about Apache Kudu?** With large number of streaming data , the load on the system increase exponential this affecting other services running on same server . **Recommendations to others considering Apache Kudu:** It is totally depends on size of streaming data source and allocation of hardware as resources. **What problems is Apache Kudu solving and how is that benefiting you?** Impala does not support processing streaming data , In order build realtime dashboard or doing OLAP operations from one query engine Apache kudu was best offering . ### 6. Apache Kudu : Review **Rating:** 4.0/5.0 stars **Reviewed by:** RAVI K. | Consultant, Enterprise (> 1000 emp.) **Reviewed Date:** November 04, 2020 **What do you like best about Apache Kudu?** Kudu-Impala Integration Features Raft Consensus Algorithm use to guarantee fault tolerance **What do you dislike about Apache Kudu?** Limited Scalability : data you can have in Kudu per cluster node and per cluster **Recommendations to others considering Apache Kudu:** For my use case Apache Kudu went well compared to others. I would suggest to start with a POC with yours used case and compare with other approach. **What problems is Apache Kudu solving and how is that benefiting you?** creation of huge datalake and setup access to dataset via Impala for downstream users. 1. Microstrategy 2. Data Science Team ### 7. Very effective for Telco data **Rating:** 4.0/5.0 stars **Reviewed by:** Verified User in Telecommunications | Mid-Market (51-1000 emp.) **Reviewed Date:** November 05, 2020 **What do you like best about Apache Kudu?** The best benefit is that you can run analytics on top of Hadoop. You don't need to propose an alternative big data solution to customer to run analytics if they have already selected Hadoop for Big Data storage. **What do you dislike about Apache Kudu?** Kudu has limitations in the type of data it can handle. You can't store complex structured data as you can do with Impala. **What problems is Apache Kudu solving and how is that benefiting you?** In my project, I had hundreds TB on Hadoop, inserted from TLC Network Nodes, monitoring systems,.. Kudu gave us the possibility to run analytics on top on our Hadoop Big and Fast Data infrastructure. ### 8. Easy SQL based tool which is easy to integrate and used for heavy workloads **Rating:** 4.0/5.0 stars **Reviewed by:** Verified User in Information Technology and Services | Enterprise (> 1000 emp.) **Reviewed Date:** December 01, 2020 **What do you like best about Apache Kudu?** Easy to use Updation and deletion which is not available in hive tables Fast insertion deletion all ACID operations **What do you dislike about Apache Kudu?** Documentations are not available specially detailing heavy workloads, optimizations **Recommendations to others considering Apache Kudu:** More good level documentations, You tube videos for implementing heavy workloads **What problems is Apache Kudu solving and how is that benefiting you?** Fast ACID transactions Easy to use SQL based Heavy workload efficient ### 9. Apache Kudu :lightning fast read and write **Rating:** 5.0/5.0 stars **Reviewed by:** Divyam G. | Big Data Developer, Enterprise (> 1000 emp.) **Reviewed Date:** April 01, 2020 **What do you like best about Apache Kudu?** no need to learn HBase for Real-Time processing, Apache Kudu can help you. easy to learn, efficient and you can update the records. it provides **What do you dislike about Apache Kudu?** Primary must be the first column, some time it is difficult to maintain 1st column as the primary key. **Recommendations to others considering Apache Kudu:** Everyone who deals with the real-time application and it will work in Transactions also , you can leverage KUDU if you or god in SQL. **What problems is Apache Kudu solving and how is that benefiting you?** while working with the Spark Streaming application kudu provides compatibility. writing is very easy and fast ### 10. Good But Consuming **Rating:** 5.0/5.0 stars **Reviewed by:** Hossam E. | Software Engineer, Mid-Market (51-1000 emp.) **Reviewed Date:** September 11, 2020 **What do you like best about Apache Kudu?** the query is very fast and has a great performance. **What do you dislike about Apache Kudu?** it consume a lot of memory which destroy the system in long time. **What problems is Apache Kudu solving and how is that benefiting you?** I face problem in 1 tablet it was full memory, I solve it by load balance. ### 11. Fast paced analytics of fast data **Rating:** 5.0/5.0 stars **Reviewed by:** Mohammedansar T. | Hadoop Administrator, Mid-Market (51-1000 emp.) **Reviewed Date:** April 01, 2020 **What do you like best about Apache Kudu?** The most useful thing is It is fully distributed and fault tolerant which makes our life easy. **What do you dislike about Apache Kudu?** The only thing I dislike is the use of primary keys, auto generated primary keys are always not supported **Recommendations to others considering Apache Kudu:** As it is fast paced data store to do some analytics on fast moving data. This is one of those which provides you reliable and required results. So, any company that wants a fast and reliable results must use this. **What problems is Apache Kudu solving and how is that benefiting you?** We are able to solve our geo fencing on data in kudu with spark. It's being quite helpful. ### 12. Best colum oriented framework **Rating:** 3.5/5.0 stars **Reviewed by:** Verified User in Marketing and Advertising | Mid-Market (51-1000 emp.) **Reviewed Date:** September 12, 2020 **What do you like best about Apache Kudu?** It's integration with Hadoop components and blazing speed.i specially like it's integration with impala **What do you dislike about Apache Kudu?** Sometime it fails execution of queries and gets stuck for indefinite amount of time. **What problems is Apache Kudu solving and how is that benefiting you?** We use kudu for analytics and data processing. Our reporting got into near real time. ### 13. Apache Kudu the right idea **Rating:** 2.5/5.0 stars **Reviewed by:** Verified User in Consumer Electronics | Enterprise (> 1000 emp.) **Reviewed Date:** June 15, 2018 **What do you like best about Apache Kudu?** Kudu is a big improvement for Impala data that needs to be quickly queried ad hoc while still being able to incrementally add to that data. **What do you dislike about Apache Kudu?** Kudu seems to have limited use in Impala. I have tried using is in Spark but didn't have great results (some of that was my fault). Impala isn't really for streaming data where the speed comes in so I'm not sure Kudu fits my use case. **What problems is Apache Kudu solving and how is that benefiting you?** Streaming data into Kudu to quickly query that data from Impala or Spark. Impala for ad hoc queries and Spark for batch processing of data. We saw significant improvements in Kudu but not that same improvements for Spark so we are not using Kudu for production work loads as ad hoc querying was not as important for us. - [View Apache Kudu pricing details and edition comparison](https://www.g2.com/products/apache-kudu/reviews?section=pricing&secure%5Bexpires_at%5D=2026-06-11+21%3A47%3A46+-0500&secure%5Bsession_id%5D=02850d5d-0966-4830-8725-2c3dc053801f&secure%5Btoken%5D=5e7d27bc00199fefcd2eb797eba36c78f955387be2b978d852b45f9a74c0f7cb&format=llm_user) ## Apache Kudu Features **Storage** - Data Model - Data Types **Availability** - Auto Sharding - Auto Recovery - Data Replication **Performance** - Integrated Cache **Security** - Role-Based Authorization - Authentication - Audit Logs - Encryption **Support** - Multi-Model - Operating Systems ## Top Apache Kudu Alternatives - [Snowflake](https://www.g2.com/products/snowflake/reviews) - 4.5/5.0 (705 reviews) - [ClickHouse](https://www.g2.com/products/clickhouse/reviews) - 4.5/5.0 (22 reviews) - [Google Cloud BigQuery](https://www.g2.com/products/google-cloud-bigquery/reviews) - 4.5/5.0 (1,148 reviews)