Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Database Benchmarks Lie (If You Let Them)

exasol.com

Post reply on HN
Print view View original

11–13 of 13 posts

Re: Database Benchmarks Lie (If You Let Them)

#11

Post by exagolo »

asteroidtunnel wrote: post #10

When I search for "high performance analytical database" in Bing, Ai summarized results are ClickHouse, Apache Druid, Singlestore, Couchbase, and Apache Pinot are considered among the best databases for real-time analytics due to their low query latency and high performance. In Google, Ai summarized results are ClickHouse, StarRocks, Snowflake, and Google BigQuery. Clickhouse is there in both of them and Exasol is no…

In case you have a single table with time-series data, then Clickhouse will perform typically better. It's very much optimized for this type of use cases. Once you are joining tables and having more advanced analytics, than Exasol will easily outperform it.

Exasol has been performance leader for more than 15 years in the market, as you can see in the official TPC-H publications, but has not gotten the broader market attention yet. We are trying to change that now and have recently been more active in the developer communities. We also just launched a completely free Exasol Personal edition that can be used for production use cases.

Re: Database Benchmarks Lie (If You Let Them)

#12

Post by ugamarkj »

asteroidtunnel wrote: post #9
ugamarkj wrote: post #8

This got me curious about our Exasol environment, which we've been running since 2016 at Piedmont Healthcare. We average 2 million queries per day (DDL/DML/DQL). Our query failure rate is ~0.1%. Only 7% of those failures were due to hitting resource limits. The rest were SQL issues: constraint errors, data type issues, etc. Average connected users is ~400. Average concurrent queries is ~7 with a daily max average of…

Very interesting. What are the bottlenecks you've faced with Exasol? "200k values in a WHERE clause IN statement"? What is that column about? Average concurrent query is ~7 in what time period?

As far as bottlenecks, I haven't really hit any in the 10 years we've been using it. Any bottleneck pain points are really user induced. We had some initial system instability during our upgrade from v7 to v8, which was a significant platform architecture change. Those issues have now been resolved. Exasol has pretty good support.

Regarding the 200k values in a where clause, we have some users that will do research across published data source in Tableau. They will copy account IDs from one report and paste them into a filter in another. Our connections from Tableau to Exasol are live. Tableau doesn't have great guardrails on the SQL that gets issued to the database.

The concurrent query comes from a daily statistics table in Exasol. There is an average and max concurrency measure aggregated per day. I averaged the last 30 days. Exasol doesn't really explain their sampling methodology in their documentation: https://docs.exasol.com/db/latest/sql_references/system_tabl...

Re: Database Benchmarks Lie (If You Let Them)

#13

Post by dashdoesdata »

asteroidtunnel wrote: post #10

When I search for "high performance analytical database" in Bing, Ai summarized results are ClickHouse, Apache Druid, Singlestore, Couchbase, and Apache Pinot are considered among the best databases for real-time analytics due to their low query latency and high performance. In Google, Ai summarized results are ClickHouse, StarRocks, Snowflake, and Google BigQuery. Clickhouse is there in both of them and Exasol is no…

You need to look at use-case alignment as well as performance.

Apache Pinot, Druid and Clickhouse are designed for low-latency analytical queries at high concurrency with continuous ingestion. Pinot is popular because of it's native integration with streaming systems like Kafka, varied indexing, and it's ability to scale efficiently. They're widely used in observability and user-facing analytics – which are how “real-time analytics databases” are commonly perceived today.

Exasol (and SingleStore, Snowflake, BigQuery, etc) are more focused on enterprise BI and complex SQL analytics rather than application serving, or ultra-high ingest workloads. It performs well for structured analytical queries and joins, but it’s less commonly deployed with the user-facing analytics or high volume usage.

A good rundown from Tim Berglund in this video here: https://startree.ai/resources/what-is-real-time-analytics/

Post reply on HN