Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>Cassandra is perfectly able to handle <em>tens of thousands</em> small writes per second on a single node. I just checked on my <em>laptop</em> and got about 29000 writes/second from cassandra-stress on Cassandra 1.2. So 500 writes per second is not really an impressive number even for a single node. </p> <p>However beware that there is also a limit on how fast data can be flushed to disk and you definitely don't want your incoming data rate to be close to the physical capabilities of your HDDs. Therefore 500 writes per second can be too much, if those writes are big enough.</p> <p>So first - what is the average size of the write? What is your replication factor? Multiply number of writes by replication factor and by average write size - then you'll approximately know what is required write throughput of a cluster. But you should take some safety margin for other I/O related tasks like compaction. There are various benchmarks on the Internet telling a single m1.xlarge instance should be able to write anywhere between 20 MB/s to 100 MB/s...</p> <p>If your cluster has sufficient I/O throughput (e.g. 3x more than needed), yet you observe OOM problems, you should try to:</p> <ol> <li>reduce memtable_total_space_mb (this will cause C* to flush smaller memtables, more often, freeing heap earlier)</li> <li>lower write_request_timeout to e.g. 2 seconds instead of 10 (if you have big writes, you don't want to keep too many of them in the incoming queues, which reside on the heap)</li> <li>turn off row_cache (if you ever enabled it)</li> <li>lower size of the key_cache</li> <li>consider upgrading to Cassandra 2.0, which moved quite a lot of things off-heap (e.g. bloom filters and index-summaries); this is especially important if you just store lots of data per node</li> <li>add more HDDs and set multiple data directories, to improve flush performance</li> <li>set larger new generation size; I usually set it to about 800M for a 6 GB heap, to avoid pressure on the tenured gen. </li> <li>if you're sure memtable flushing lags behind, make sure sstable compression is enabled - this will reduce amount of data physically saved to disk, at the cost of additional CPU cycles</li> </ol>
    singulars
    1. This table or related slice is empty.
    1. POHow much load can cassandra handle on m1.xlarge instance?
      singulars
      1. PTQuestion
    1. PTAnswer
    1. This table or related slice is empty.
    1. USPiotr Kołaczkowski
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POHow much load can cassandra handle on m1.xlarge instance?
      singulars
      1. PTQuestion
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTAcceptedByOriginator
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COWell for the rate of our I/O, performance was awful (even with ephemeral storage), so we ended up ditching it. I don't think we were pulling a lot of data into it as we were reading/writing data from S3 and dumping it into C* (Although it was realtime). There could have been a parameter that could have improved our performance but in the end we decided not to waste our time. Don't get me wrong, it may work in a lot cases and some people are big fans, but it didn't work us. It's our strictly subjective opinion.
      singulars
      1. PO
      1. USRico
    2. COI further investigated the cluster. I run number of stress tests on it (including the one attached to datastax java drivers) and it looks like the cluster itself is able to manage about 5K writes per second during the tests. It also behaves stable. Now I think that the problem is in the specific data structure that we have, or maybe driver usage? Is it possible that composite keys have such a performance penalty? What else may matter?
      singulars
      1. PO
      1. USBartek
    3. CO@Rico which version was it? Maybe you were hit by some bug? E.g. just recently I stress-tested an early C* 2.0 version and got awful performance, then realized I'm not on the latest stable version - upgraded and it worked like charm.
      singulars
      1. PO
      1. USPiotr Kołaczkowski
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload