Re: Performance comparisons
"makes sense to run benchmarks" and "makes sense to force everyone to run their own" aren't the same. I think you support the first, whereas I argue with the second. Nearly everyone publishes numbers vs Cassandra which makes it easy to compare tuned setups of any two products (except Cassandra which everyone is leaving untuned;-). -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: Performance comparisons
If 3-rd party runs benchmark even better. It doesn't make sense if everyone in the world has to run the same benchmarks. Division of labor makes sense. Here's what I meant w.r.t. comparison vs Hazelcast. https://www.gridgain.com/resources/blog/gridgain-confirms-apache-ignite-performance-2x-faster-hazelcast https://blog.hazelcast.com/fake-benchmark/ -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Performance comparisons
Cannot find performance reports other than vs Hazelcast. Persistent Ignite is particularly interesting. Does anything of this sort exist? -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: Cassandra failing to ReadThrough using Cache.get(key) without preloading
Last time I tried, which was early 2017, cache.getEntry() would pull it from Cassandra (or whichever database is backing the cache) for me, while cache.get() won't. I then found somewhere, in the docs or on the board, that it's to be expected. Can someone confirm? I was most interested in SQL, which definitely won't see entries that aren't preloaded into caches, not clearly remembering "get" vs "getEntry" differences. -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: Query about running SQL with durable memory
If memory cache is empty and all 10 entries are on disk, will Ignite de-serialize all of them? Can de-serialization be avoided if the query has a filter, e.g. SELECT COUNT(*) FROM EMP WHERE SALARY > ? -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Query-about-running-SQL-with-durable-memory-tp16015p16180.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.
cache insert/update performance
Has anyone measured x86 instruction count needed by a single cache insert/update once it gets to the correct data node? Lets say for String key and value or String key and object value with a couple of scalar members. -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/cache-insert-update-performance-tp14049.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.
Re: External data sources
Hi Val, so nothing in progress and no immediate plans, is that right? -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/External-data-sources-tp11766p11859.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.
Re: External data sources
As I understand, Ignite has two completely disconnected features - write-through/read-through caching and SQL. If SQL fetched what's not currently cached in memory from the cache store that would solve the problem. This is only one possible solution and it covers only the sources described/cached by Ignite caches. It's also possible to add an API to expose external sources to SQL without threading it through an Ignite cache (something like external rdbms tables, Presto plugin, etc.). Any plans to implement either of the two or anything equivalent? Thanks. -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/External-data-sources-tp11766p11829.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.
Re: External data sources
Thanks, Andrey, I know it doesn't just work out of the box, but what's the best way to add something like that? It's a fairly common feature, Oracle has external tables, Presto has a plug-in mechanism that allows, among other things, to pushdown predicates to the external table driver. -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/External-data-sources-tp11766p11786.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.
External data sources
What's the best way to implement SQL that joins an Ignite cache with an external data source, e.g. a Cassandra column family or a MySql table? A very big external table, not one that can be cached in memory prior to running the query. -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/External-data-sources-tp11766.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.