A Brief Introduction to MemSQL

Jenifer Ho
Jenifer Ho

We know choosing or evaluating a new database technology can be challenging due to the variety of choices available. In a recent webcast, we shared various use cases businesses face with traditional database and data warehouse technologies, key differentiators and architectures of MemSQL, sample applications and customer case studies, and a quick demo of MemSQL.

MemSQL provides an adaptable database for real-time applications that unite transactions and analytics in a single high-performance platform. MemSQL uniquely delivers fast data ingestion with low latency while serving large numbers of simultaneous users.

Here are common answers to questions from the webcast:

Question: Is there an option to leverage in-memory for the column format OLAP queries?
Answer: Yes, for optimal performance, MemSQL leverages both memory and disk for the columnstore engine.

Question: Does MemSQL require partitioning of data for distributed ACID consistency?
Answer: MemSQL automatically partitions data data, so this is built in.

Question: Is there any IDE to use with MemSQL?
Answer: MemSQL supports the MySQL wire protocol and works with many development IDEs including Sequel Pro, TOAD, Navicat, and MySQL Workbench to name a few.

Question: Can MemSQL replace Teradata workloads and help us reduce cost?
Answer: Yes. MemSQL is a distributed relational SQL database and data warehouse in one. The columnstore engine can query up to petabytes of data while providing a SQL end point to business intelligence tools, data scientists, and data analysts. MemSQL makes use of industry standard hardware for optimal cost performance compared to traditional data warehouse appliances.

Question: Do you have use cases where users are storing large numbers of large images (tiff/jpg/png) such as microscopy or satellite imaging applications?
Answer: While it is possible to store binary objects (such as images) in MemSQL, many customers use object stores like S3 for this purpose. However, the metadata associated with those images, and the ability to rapidly scan that structured metadata would be a good use case for MemSQL.

Question: Is data automatically synced between row store and column store or does the DBA have to manually manage that transfer?
Answer: Data is automatically placed in the table type or storage engine you choose. From there the DBA can move data as needed. You can learn more about how our storage engines work at our documentation page here.

Question: Is the MemSQL shell using MySQL CLI or do you have your own CLI for BI access? Is there something other than ODBC?
Answer: We are currently building a MemSQL CLI environment. Stay tuned for our next release! MemSQL supports JDBC and ODBC connectivity.

Question: Is there a free developer version, perhaps limited size or functionality?
Answer: We do offer a free developer edition with unlimited scale, although it does not include security or redundancy. Learn more about signing up at

Question: Are the MemSQL features also supported for Azure as well?
Answer: Yes, MemSQL can run on any cloud platform, or on-premises.

Question: How are encryption keys managed? Are cryptographic hardware security modules (HSM) supported?
Answer: We find it best for customers to manage their own encryption keys.

Question: Regarding Geospatial support, is the data we would load into MemSQL in GeoJSON format the way it would be in a MongoDB environment?
Answer: MemSQL supports string, number, or JSON format for Geospatial data. You can learn more about our Geospatial support here.

Question: Does MemSQL support a cloud solution?
Answer: Yes. MemSQL can be run on any public or private cloud including AWS, Microsoft Azure, and Google Cloud Platform.

Question: Can you demo the ETL part?
Answer: We would love to show you this. We’ll follow-up with you to schedule some time to demo this or book some time with one of our product experts at

Question: What are the minimum memory requirements?
Answer: For small development or prototyping, you can run MemSQL with 8GB of memory. Our hardware requirements are posted here.

Missed the webcast? Get a copy of the recording to watch on demand here.

To talk more in depth about your database challenges, please schedule time with a product expert at



MemSQL Helios eclipse
MemSQL Helios
The World’s Fastest Cloud Database