Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Accumulo, mail # user - tservers running out of heap space


Copy link to this message
-
Re: tservers running out of heap space
Anthony Fox 2012-11-29, 19:24
Since the scan involves an intersecting iterator, it has to scan the entire
row range.  Also, it's not even very many concurrent clients - between 5
and 10.  Should I turn compression off on this table or is that bad idea in
general?
On Thu, Nov 29, 2012 at 2:22 PM, Keith Turner <[EMAIL PROTECTED]> wrote:

>
>
> On Thu, Nov 29, 2012 at 2:09 PM, Anthony Fox <[EMAIL PROTECTED]> wrote:
>
>> We're not on 1.4 yet, unfortunately.  Are there any config params I can
>> tweak to manipulate the compressor pool?
>
>
> Not that I know of, but its been a while since I looked at that.
>
>
>>
>>
>> On Thu, Nov 29, 2012 at 1:49 PM, Keith Turner <[EMAIL PROTECTED]> wrote:
>>
>>>
>>>
>>> On Thu, Nov 29, 2012 at 12:20 PM, Anthony Fox <[EMAIL PROTECTED]>wrote:
>>>
>>>> Compacting down to a single file is not feasible - there's about 70G in
>>>> 255 tablets across 15 tablet servers.  Is there another way to tune the
>>>> compressor pool or another mechanism to verify that this is the issue?
>>>
>>>
>>> I suppose another way to test this would be to run a lot of concurrent
>>> scans, but not enough to kill the tserver.  Then get a heap dump of the
>>> tserver and see if it contains a lot of 128k or 256k (can not remember
>>> exact size) byte arrays that are referenced by the compressor pool.
>>>
>>>
>>>>
>>>>
>>>> On Thu, Nov 29, 2012 at 12:09 PM, Keith Turner <[EMAIL PROTECTED]>wrote:
>>>>
>>>>>
>>>>>
>>>>> On Thu, Nov 29, 2012 at 11:14 AM, Anthony Fox <[EMAIL PROTECTED]>wrote:
>>>>>
>>>>>> I am experiencing some issues running multiple parallel scans against
>>>>>> Accumulo.  Running single scans works just fine but when I ramp up the
>>>>>> number of simultaneous clients, my tablet servers die due to running out of
>>>>>> heap space.  I've tried raising max heap to 4G which should be more than
>>>>>> enough but I still see this error.  I've tried with
>>>>>> table.cache.block.enable=false
>>>>>> table.cache.index.enable=false, and table.scan.cache.enable=false and
>>>>>> all combinations of caching enabled as well.
>>>>>>
>>>>>> My scans involve a custom intersecting iterator that maintains no
>>>>>> more state than the top key and value.  The scans also do a bit of
>>>>>> aggregation on column qualifiers but the result is small and the number of
>>>>>> returned entries is only in the dozens.  The size of each returned value is
>>>>>> only around 500 bytes.
>>>>>>
>>>>>> Any ideas why this may be happening or where to look for further info?
>>>>>>
>>>>>
>>>>> One know issues is hadoops compressor pool.  If you have a tablet with
>>>>> 8 files and you query 10 terms, you will allocate 80 decompressors.   Each
>>>>> decompressor uses 128K.   If you have 10 concurrent queries, 10 terms, and
>>>>> 10 files then you will allocate 1000 decompressors.    These decompressors
>>>>> come from a pool that never shrinks.  So if you allocate 1000 at the same
>>>>> time, they will stay around.
>>>>>
>>>>> Try compacting your table down to one file and rerun your query just
>>>>> to see if that helps.   If it does, then thats an important clue.
>>>>>
>>>>>
>>>>>
>>>>>>
>>>>>> Thanks,
>>>>>> Anthony
>>>>>>
>>>>>
>>>>>
>>>>
>>>
>>
>