SingleStore

Formerly MemSQL

We found SingleStore to be significantly superior to bare-bones MySQL as it offers better performance and scalability. It is also richer in terms of the query language supported, stored procedures, and connectivity. We also found it to be more robust than MySQL and with better business support.

SingleStore is built for fast data ingestion and fast queries against large tables (> billions of rows). This is possible because of the column store engine that SingleStore uses. SingleStore also support a memory engine. Pipelines is also another big advantage. Being able to ingest data from S3 (and loading new data) without adding any additional services (with just 1 SQL query) is pretty cool. SingleStore is also pushing hard with vector engine. It is becoming a single storage solution.

SingleStore has outperformed these in speed and performance. However it is more expensive but has been worth the cost so far.

We knew early on that MySQL (Amazon Aurora) would not be suitable for this workload as it cannot query our time series data as fast as SingleStore. We also use MongoDB Atlas for another application but we could not achieve the raw speed we saw from SingleStore. Our technical staff have used SQL for many years, if we used MongoDB they would need to learn the MongoDB query and aggregation syntax.

We were initially using AWS Aurora which worked well at the time but as we grew it just felt like a lumbering beast - even with tier upgrades. We started looking at caching & search options to help make searches faster - we were already using redis. We looked at and even implemented Elasticsearch and then ended up ditching that after about six months. For our usage, it caused more problems than it helped and had a lot of overhead and cost. We then looked at Algolia which seemed rather expensive to add on top of our current db expenses but then found the opensource meilisearch. We actually liked that quite a bit but weren't sure if we wanted to have to implement, host and maintain it. It also wasn't scalable (at that time) the way we wanted for future growth. It solved some problems but not all. SingleStore solved all these problems while essentially trading the cost almost straight across from Aurora and removed all the operating overhead in terms of maintenance. It was a no brainer.

Timescale was the biggest alternative option we looked at for SingleStore, however the requirement to learn a new syntax (due to not being SQL compatible) was our biggest pain point.

Supporting a new language would require alterations to the Laravel framework, as this only offered SQL integration out of the box. This alteration would be time consuming and would limit our scope to future hiring due to the new syntax.

SingleStore provides an abstraction layer in managing a sharded database solution reducing complexity for the FLOWD team. Coupled with the SingleStore Managed Service, we are partnering with SingleStore to provide FLOWD services to various utilities & councils.

SingleStore is eons faster than other database providers, and it absolutely crushes calculations & aggregations. While other providers may have a few quality of life enhancements over SingleStore, the speed benefits of SS far outweigh the cons. At the end of the day, speed matters a lot in our workflows, and Singlestore allows us to run things in the desired timeframes.

SingleStore provides performance for real-time data analytics application where as Snowflake is more for data engineers and not fast enough for real-time data analytics apps.

As I said before, we felt that running queries on BigQuery for every query is really slow, especially from an user point of view. After seeing the drastically improved latency of SingleStore we decided to use to solve this issue. We currently use it to run low volume queries for a much faster response, significantly improving user experience.

SingleStore was the fastest sql database we have tried, allowing for analytical queries on TBs of data.

We've tried both these above, to solve ingestion and analytical queries. Dynamo was a perfect ingestion service, but it didn't allow us to run intense analytical queries. Aurora helped us run these analytical queries, and ingestion was reasonable, BUT the queries were simply too slow. SingleStore has provided a solution to essentially allow us to take these two tools, and turn them into 1.

Very similar offerings. SingleStore promise a faster query speed and better cost structure. Vector databases is anoter offering that is unique to SingleStore.

SingleStore is much faster than others in terms of data ingestion and query response time

SingleStore outperformed both MongoDB and PostgreSQL for OLAP workloads. Its ability to shard data and handle parallel processing of SQL "JOIN" queries across shards is a game changer.

We looked at Microsoft Azure SQL, MariaDB, Vertica, and RedShift. Our criteria were cost and performance. We also needed compatibility with the Azure cloud as this is where our other systems were running. This was in 2020. After approximately three months of evaluation, we chose SingleStore. SijgleStore was also very supportive during the transition from our old provider to their environment.

SingleStore is simply easier to use and most importantly much, much faster. For Big Data, it is an absolute life saver :)

We selected SingleStore due to its multi cloud nature, SQL compatibility and customer references who operate similar applications as us

It seems that SingleStore is good at being able to handle complex queries against large datasets out of the box. In the past, we've had to do quite a bit of manual configuration and database performance tuning, but SingleStore (so far) has seemed to require minimal configuration in this aspect. Both data imports/exports, as well as queries against the data, run very fast.

SingleStore achieves same or much better performance, while avoiding all data sync and migration

NA

The SQL dialect seems to be pretty compatible, and I especially appreciate that you can connect to it with native MySQL tooling.

Well honestly, Initially I thought it would be just another SQL Relational Database Management System but when I actually signed up and use SingleStore, I was amazed to see such integration of intuitive and easy-to-use UI and such powerful performance. So, the CSV for cosmetic shop data was pretty big in size(415 MB extracted) containing records in million, was shocked to see how fast the queries were running on top of the data once loaded to SingleStore, I even tried running some queries on a sample stock trade data and then running queries which involved compute-intensive operations like joins and the results were returned in no time. Also, I never had an experience before using SingleStore where I created a pipeline from S3 in seconds. Now, coming to your questions

first and important advantages is super fast, the million of data can be treated on Seconds