Glyn Heatley, Cervello’s Analytics and Information Management Co Practice Lead speaks to Marc Hassan, a Senior Consultant within the practice about this sophisticated in-memory platform.
1. So what is MemSQL?
“MemSQL is a high-performance, in-memory database that combines the horizontal scalability of distributed systems with the familiarity of SQL. Translated that means… MemSQL is a distributed, in-memory database. It is somewhat unique in that users can write plain SQL queries against it without scripting or learning a new query language (like some other in-memory databases require). Because it is a distributed database, it is highly scalable in terms of data volume and performance. Data is stored in-memory, so it requires large amounts of RAM but performs much better than a traditional database.”
2. How did you come across it/hear about it?
“I came across the press release from MemSQL which piqued my interest. Anything that is progressive and free to tinker with is interesting to a developer, especially when one is looking for new solutions to complex problems.”
3. Why is it interesting to you? To others?
“In this release, MemSQL launched a free ‘Community Edition’, which is uncommon when it comes to in-memory databases. This was a great opportunity for us to try it out and benchmark it with other in-memory databases that we have used. Another reason it is interesting is that it uses the same protocols as MySQL, which makes it very easy to integrate with.”
4. What did you do to get it up and running? And where?
“First, I created a new AWS Virtual Private Cloud. Then, I launched a 64-bit Ubuntu Linux EC2 instance. I downloaded and installed MemSQL Community Edition, and I was up and running in no time! I configured a basic environment: one Master Aggregator and one Leaf node. I was immediately able to create a new database and tables.”
5. What sample data did you load to it?
“I created some dummy data using cartesian joins of a table with itself. I ended up with ~2 million rows in a table. A few queries that I wrote performed very well, returning data in a few seconds. A similar query in a traditional RDBMS would probably take two or three times longer.”
6. Now that you’ve got it up and running what do you think about it? What do you like? How do you think it can be improved?
“It is very easy to set up, and I like that it uses the same protocol as MySQL. Also, I like that they released a Community Edition. It shows that they are forward-thinking and open.”
7. What are some other similar technologies you would compare it to?
“SAP HANA and Oracle TimesTen.”
8. How are you planning to use it? Part of a project? Part of an architecture?
“We are hoping to load some real data into it and compare its performance with other databases, notably SAP HANA. We are using SAP HANA on a few projects , but it is always good to see potential alternative as one vendor solution does not always fit all. We are interested in this platform as the ability to treat it like a MySQL database and write standard SQL against it is a huge plus.”
Do you have experience with MemSQL or are you looking to leverage it in some creative way? Leave a comment. We’d love to hear from you.
Interested in learning more? Join our mailing list!