Apache Hive vs. OpenText Vertica

Apache Hive shines for ad-hoc analysis and plugging into BI tools. Its SQL-like syntax allows for ease of use not for only for engineers but also for data analysts. Through our experience, there are probably more desirable tools to use if you are planning on integrating Hive into your processing pipeline.

Incentivized

As someone just starting out with data analytics and warehousing vertica is a great tool for a small scale business. It has amazing performance and can scale upto TBs of data. It works well for any organization which has about 100 - 500 DAUs of the system. The system doesn't require a lot of ops overhead. Scaling for PB data and 1000s of DAU is vertica's weak point. The system is just not designed for large scale usage and still has a long way to go to improve scalability. There are experiments to run Vertica query engine on top of HDFS which seem promising, however - if you have the the Hadoop ecosystem you are better off going the HDFS + Presto/Impala/SparkSQL route. But if you are in the Hadoop ecosystem, you probably are already investing a lot in ops.

Incentivized

• Hive syntax is almost like SQL, so for someone already familiar with SQL it takes almost no effort to pick up Hive.
• To be able to run map reduce jobs using json parsing and generate dynamic partitions in parquet file format.
• Simplifies your experience with Hadoop especially for non-technical/coding partners.

Incentivized

• Column-oriented storage organization, which increases performance of queries.
• Compression, which reduces storage costs and I/O bandwidth. High compression is possible because columns of homogeneous datatypes are stored together and because updates to the main store are batched.
• Shared nothing architecture, which reduces system contention for shared resources and allows gradual degradation of performance in the face of hardware failure.
• Easy to use and maintain through automated data replication, server recovery, query optimization, and storage optimization.
• Support for standard programming interfaces ODBC, JDBC, ADO.NET, and OLEDB.
• Integration to Hadoop with the capability to perform analytics on ORC and Parquet files directly.

Incentivized

• Use Hive for analytical work loads. Write once and read many scenarios. Do not prefer updates and deletes.
• Behind scenes Hive creates map reduce jobs. Hive performance is slow compared to Apache Spark.
• Map reduce writes the intermediate outputs to dial whereas Spark operates in in-memory and uses DAG.

Incentivized

• One time, one of the nodes wasn't coming up because of some ambiguity with the local data. Vertica wasn't able to fix it by itself and we were trying to remove the node out of the database and we couldn't do it. It would be great if that could be addressed. Luckily when we rebooted the whole server, some of the dead transaction got flushed because of which vertica was able to recover and the node came up.

Incentivized

Since I do not know the second data warehouse solution that integrate with HDFS as well as Hive.

Hive is a very good big data analysis and ad-hoc query platform, which supports scaling also. The BI processes can be easily integrated with Hadoop via the Hive. It can deal with a much larger data set that traditional RDBMS can not. It is a "must-have" component of the big data domain.

Incentivized

Apache Hive is a FOSS project and its open source. We need not definitely comment on anything about the support of open source and its developer community. But, it has got tremendous developer support, awesome documentation. I would justify the fact that much support can be gathered from the community backup.

Incentivized

HP/Micro Focus Vertica support is in par with other bigger vendors. In addition to this, there is enough best practices documentation available for some of the most common ways you will use Vertica that makes it easy to get Vertica up and running.

Incentivized

We have used a simple but necessary function such as merging certain data tables, which although they may be from different areas, complement each other or are necessary, you can use metadata if what you need is to validate the origin of your information and what impact it has, is also feasible.

Incentivized

MySQL and MS SQL Server are both fantastic RDBMS products. MS SQL Server goes a bit further since it has the builtin analytical functions. But it only scales so far. Once the data goes beyond capacity, getting results out just does not happen anymore. IBM Netezza and Teradata were both appliances that required different expertise than we had in house. Vertica was able to do the same, and in some cases better, on commodity hardware (frankly in our case old servers that were slated for recycling!) and at a small scale. In other words, Vertica we could grow slowly over time. Infobright is a great log processing database but for the functions we were looking to serve it just didn't have some of the features Vertica had that we felt were show stoppers.

Incentivized

• Good ROI for being able to access data easily across the network, we have large amounts of data and this is a good system to access it
• Good ROI for being easy to learn how to use for new employees, not much time spent which saves costs
• Good ROI for being able to integrate with Spark and other applications, hence data can be analyzed through programs

Incentivized

• Vertica increased our productivity in analyzing the data and validating simple proof of concepts with our data.
• Results of analytical queries produced from Vertica are used by all departments as well as part of some of our products.

Incentivized