On 11/12/13, 1:25 PM, Terry P. wrote:
Hi Josh, Thanks for your exhaustive reply. I am using Native maps, and it's set to 1G in my accumulo-site.xml. The data and index cache settings there are still at their 3G default values as well (50M and 512M). I definitely didn't realize that and will increase their size given I have plenty of memory sitting around idle (it was intended to be used for caching too!). Will increasing the tserver.memory.maps.max in accumulo-site.xml perhaps help reduce these warning messages? My only concern is that an operator may be monitoring the Accumulo Monitor GUI and see the memory warnings and think "Oh no, we're almost out of memory, I should page someone!"
Hahaha, yeah, I know what you mean. You can tell them that it's just a "warning" and not an "error" :P. Increasing the size of the memory maps won't make the error go away. I believe that warning is purely over JVM heap. I don't believe there's any code (outside of the flush-policy for when to close a native map and start a new one to make sure the tserver.memory.maps.max is observed) to monitor the size of the native maps.
You would want to increase JVM heap size to keep that error from happening or reduce the amounts of heap you give to the index block or data block cache.
Based on what you've seen, is the warning innocuous and can just be ignored?
IMO, yes. Given with a strong recommendation that you know that you're not spending any significant time in garbage collection.
`fgrep 'gc ParNew'` on your tserver.debug.log and not seeing "spiky" gc cycles.
On Tue, Nov 12, 2013 at 3:03 PM, Josh Elser <[email protected] <mailto:[email protected]>> wrote: IMO, I see this at home on my computer no matter what memory settings I use. I've become pretty accustomed to flat out ignoring it... As for heap management, there are two big paths here: with "native maps" and without. When you write data to Accumulo, it goes to two places: 1) Write-ahead log and 2) Memory maps. The WAL ensures that if you have writes in memory on a server that dies, that you don't lose data. The memory maps give you much faster ingest over trying to write into a sorted file. 1) Native maps (aka c++ code over JNI) This memory allocation, controlled by tserver.memory.maps.max in accumulo-site.xml, is "off heap" memory. It is not limited by the JVM heap limits you specify in ACCUMULO_TSERVER_OPTS in accumulo-env.sh. As such, you need to make sure that you don't over-allocate memory usage on your node (tserver.memory.maps.max + JVM Xmx + fudge-factor < total available memory). 2) Non-native (in JVM) This serves the same purpose as #1 but is in JVM heap as opposed to off heap. Ingest will be slower and JVM gc will likely be a bigger issue than using the native maps. This does make the JVM sizing a little more straightforward: JVM Xmx + fudge-factor < total available memory (but math is pretty easy). Assuming you use the native maps, lets break down what you see in JVM heap. 1) Index block cache Each RFile (backing file for tablets in Accumulo), has an multi-level index structure which lets you efficiently find the data in that file. Accumulo provides the ability to cache this index information instead of reading and deserializing from disk every time. Controlled by tserver.cache.index.size. 2) Data block cache Similar to #1 except it's for the actual blocks of data in that RFile (the key-value pairs) instead of just the index structure. Controlled by tserver.cache.data.size. This can give you some benefit over having to hit a (potentially, remote) datanode every time you perform a read in a read-heavy environment. 3) "The rest" Consider this the rest of the things that the tabletserver does. "hosting" its tablets (each tablet has a collection of files in hdfs), scansessions running against those hosted tablets (the iterator stack that is created to perform a "read"). Compression (de)allocators for Hadoop (assuming you're using GZIP). Various internal buffers for caching. Connection management information (thrift and hadoop connections). I'm probably missing more things, too. On 11/12/13, 12:32 PM, Terry P. wrote: On an Accumulo 1.4.2 I've gotten "[tabletserver.TabletServer] WARN: Running low on memory" 5 times in the last two days on just one of my 6 datanodes. That datanode is hosting ~30% of the data, as 2 datanodes had dropped from the cluster due to a network issue some time ago and hasn't entirely rebalanced. Current volume is only 140 million. Ingest rates has been pretty constant at a light 200 per second. Not knowing how Accumulo uses its java heap space, I opted to start with a stock memory config and used the 3GB example config files, which I see allocates only 1GB to the TabletServer. The server has 24GB RAM and currently is using only 10GB total between Accumulo and HDFS, so there's plenty of free memory to spare. Is ACCUMULO_TSERVER_OPTS in accumulo-env.sh the tunable I should target to alleviate these warnings? Unfortunately being ops-configured nodes, no JDKs are installed nor is it a possibility to do so in order to monitor the JVM itself for better information. Thanks in advance, Terry
