Just a little background on the CERN results, since I’m not sure it’s clear in their presentation - they were using a very old version of InfluxDB (either v0.8 or v0.9), and it had very different (read: underwhelming) performance characteristics back then. We’ve moved to an all-new storage engine (TSM) since then, among many other significant performance and stability changes. It’s definitely safe to say that InfluxDB outperforms Elasticsearch for time series data now.
I don’t have any experience with using Aerospike for time series, so I can’t speak to the performance comparison there. Maybe someone else on the list can? Todd On Mon, Jan 9, 2017 at 4:19 PM, <[email protected]> wrote: > In working on a data ingestion platform, one of the requirements is a > low-latency real-time ML-related response using Kafka Streams. > > Kafka is used as the message broker, and as of v0.10, Kafka Streams will > be used in the place of something like Flink or Spark. Given the TICK > stack, how would Telegraph fit in with Kafka Connect? > > The initial database to use was ElasticSearch, as another one of the > requirements was to have text proximity search results and NLP. However, > since that is a completely separate use-case from time-series data, we'll > stick with the real-time requirement and a separate database for this use > case: > > Aerospike > InfluxDB > OpenTSDB > > This guy at CERN http://cds.cern.ch/record/2011172/files/LHCb-TALK-2015- > 060.pdf determined that ElasticSearch was the clear winner over InfluxDB > for time-series data. > > However reviewing Influx's results https://www.influxdata.com/ > influxdb-markedly-elasticsearch-in-time-series-data-metrics-benchmark/, > it is the clear winner over ElasticSearch. > > I'm sure I could find other contradicting results for anything else (but > really what was missed in the testing?). > > I would like to know from those who have actually implemented these > systems what the experience-based winner is, more specifically, between > Aerospike and InfluxDB. > > -- > Remember to include the version number! > --- > You received this message because you are subscribed to the Google Groups > "InfluxData" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/influxdb. > To view this discussion on the web visit https://groups.google.com/d/ > msgid/influxdb/19f67d62-b3d4-4dd9-ad43-7d7edca124c1%40googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Remember to include the version number! --- You received this message because you are subscribed to the Google Groups "InfluxData" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/influxdb. To view this discussion on the web visit https://groups.google.com/d/msgid/influxdb/CAKozqAqKWoi25cMdex-p3x3cE-nh2t-P729Lua5hupP%2Bbdvqkg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
