Exasol, from the company of the same name in Nuremberg, is presented by the vendor as a high-performance in-memory analytics database that aims to transform how organizations works with data, on-premises, in the cloud or both.
N/A
SingleStore
Score 9.8 out of 10
N/A
SingleStore aims to deliver the world’s fastest distributed SQL database for data-intensive applications: SingleStoreDB, which combines transactional + analytical workloads in a single platform.
We looked at some others too, but was 5 yrs ago so I don't recall the list. Exasol had the best performance per cost, outstanding performance, and was easy to evaluate. Even their community addition running on my laptop was faster than our existing reporting solution.
Exasol is suited to applications requiring fewer & larger queries (reporting/data analytics/business intelligence tools). Its per-query overhead makes it unsuitable as an operational database (those are optimized for many & smaller queries.)
SingleStore HTAP engine is well suited for real-time analytics, fast ingestion, scaling OLTP system like MySQL. When you need to run reports or perform aggregates on billions of rows and you get result in seconds, you cannot get this experience with other OLTP engines. I wish DBtLab was a little more developer and supported for SingleStore. This would allow to perform better data transformation. You can use stored procedures, but DBTLabs has become a standard for dimensional modeling in data warehousing projects. This is probably why SingleStore has trouble piercing in the data warehouse world. It is definately capable to compete with Snowflake when it comes to scalability, query performance, data compression, but Snowflake has ravaged the data warehouse market in few years and large corporations have already invested lots of money in migrating into Snowflake. The SingleStore community needs to grow. Everyone who uses SingleStore loves it.
We have found Exasol to be very fast at summarizing large data sets. It has been a great backend for both reporting tools and data analytics/business intelligence. Combined with the fact that data import is also very fast it makes it ideal for a real-time ELT architecture.
Exasol is low maintenance. No indexes to maintain (The database auto-manages them) and very little tuning is required.
Query processing is optimized for high throughput and high parallelization. This means that even under high loads performance degrades gracefully as opposed to having "pile-ups" and "meltdowns". This has made it a very reliable database for us.
Exasol doesn't have some of the advanced enterprise-y features found in some other large corporate database systems (e.g. native row-level and column-level security). However it's pretty customizable (SQL pre-processing, virtual schemas, powerful user-defined-function frameworks) so many of these features can be implemented manually.
We wish the product had better support for High Availability of the aggregator. Currently the indexes generated by the two different aggregators are not in the same sequential space and so our apps have more burden to deal with HA.
More tools for debugging issues such as high memory usage would be good.
The price was the one that kept us away from purchasing for the first few years. Now we are able to afford due to a promotion that gives it at 25% of the list price. Not sure if we'll continue after the promotion offer expires in another 2 years.
I gave it 9/10 instead of 10/10 only because it lacks in a few advanced enterprise features that require manual workarounds. Otherwise our users have had no problem getting up to speed with it (other than SQL syntax issues that are specific to it, but that's true of any DB)
[Until it is] supported on AWS ECS containers, I will reserve a higher rating for SingleStore. Right now it works well on EC2 and serves our current purpose, [but] would look forward to seeing SingleStore respond to our urge of feature in a shorter time period with high quality and security.
SingleStore's performance is incredible. Our predictive algorithms went from taking 24-48 HOURS down to 15 minutes allowing our team to run those much more frequently. Previously, we were limited to about 60 requests per minute due to table locks. Implementing columnstore on SingleStore allowed us to receive 1000 requests per minute.
I have had only positive experiences with their support. They are fast, knowledgeable, and courteous. Online support requests get picked up within hours. I've only once had to use their hotline and that was for an emergency. There was even one minor non-security bug report that I reported and which they fixed in the following week's minor release. I was quite impressed.
Very responsive to trouble tickets - Often, I think, the SingleStore's monitoring systems have already alerted the engineers by the time I get around to writing a ticket (about 10 - 20 mins after we see a problem). I feel like things are escalated nicely and SingleStore takes resolving trouble tickets seriously. Also SingleStore follows up after incidents to with a post mortem and actionable takaways to improve the product. Very satisfied here.
We allowed 2-3 months for a thorough evaluation. We saw pretty quickly that we were likely to pick SingleStore, so we ported some of our stored procedures to SingleStore in order to take a deeper look. Two SingleStore people worked closely with us to ensure that we did not have any blocking problems. It all went remarkably smoothly.
We looked at some others too, but was 5 yrs ago so I don't recall the list. Exasol had the best performance per cost, outstanding performance, and was easy to evaluate. Even their community addition running on my laptop was faster than our existing reporting solution.
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.
When we first migrated to Exasol (from a MySQL-backed reporting tool) our clients spent over an order of magnitude less time waiting on reports and since then it has opened up possibilities for reporting and analyzing data that were simply not possible beforehand.
As the overall performance and functionality were expanded, we are able to deliver our data much faster than before, which increases the demand for data.
Metadata is available in the platform by default, like metadata on the pipelines. Also, the information schema has lots of metadata, making it easy to load our assets to the data catalog.
