Re: Cassandra versus HBase performance study
Hi Brian. was there any performance changes on the other tests with v0.5 ? the graphs on the other pages looks remarkably identical. On Feb 4, 2010, at 11:45 AM, Brian Frank Cooper wrote: 0.5 does seem to be significantly faster - the latency is better and it provides significantly more throughput. I'm updating my charts with new values now. One thing that is puzzling is the scan performance. The scan experiment is to scan between 1-100 records on each request. My 6 node Cassandra cluster is only getting up to about 230 operations/sec, compared to 1400 ops/sec for other systems. The latency is quite a bit higher. A chart with these results is here: http://www.brianfrankcooper.net/pubs/scans.png Is this the expected performance? I'm using the OrderPreservingPartitioner with InitialToken values that should evenly partition the data (and the amount of data in /var/cassandra/data is about the same on all servers). I'm using get_range_slice() from Java (code snippet below). At the max throughput (230 ops/sec), when latency is over 1.2 sec, CPU usage varies from ~5% to ~72% on different boxes. Disk busy varies from 60% to 90% (and the machine with the busiest disk is not the one with highest CPU usage.) Network utilization (eth0 %util both in and out) varies from 15%-40% on different boxes. So clearly there is some imbalance (and the workload itself is skewed via a Zipfian distribution) but I'm surprised that the latencies are so high even in this case. Code snippet - fields is a SetString listing the columns I want; recordcount is the number of records to return. SlicePredicate predicate; if (fields==null) { predicate = new SlicePredicate(null,new SliceRange(new byte[0], new byte[0],false,100)); } else { Vectorbyte[] fieldlist=new Vectorbyte[](); for (String s : fields) { fieldlist.add(s.getBytes(UTF-8)); } predicate = new SlicePredicate(fieldlist,null); } ColumnParent parent = new ColumnParent(data, null); ListKeySlice results = client.get_range_slice(table,parent,predicate,startkey,,recordcount,ConsistencyLevel.ONE); Thanks! Brian From: Brian Frank Cooper Sent: Saturday, January 30, 2010 7:56 AM To: cassandra-user@incubator.apache.org Subject: RE: Cassandra versus HBase performance study Good idea, we'll benchmark 0.5 next. brian -Original Message- From: Jonathan Ellis [mailto:jbel...@gmail.com] Sent: Friday, January 29, 2010 1:13 PM To: cassandra-user@incubator.apache.org Subject: Re: Cassandra versus HBase performance study Thanks for posting your results; it is an interesting read and we are pleased to beat HBase in most workloads. :) Since you originally benchmarked 0.4.2, you might be interested in the speed gains in 0.5. A couple graphs here: http://spyced.blogspot.com/2010/01/cassandra-05.html 0.6 (beta in a few weeks?) is looking even better. :) -Jonathan -- Ian Holsman i...@holsman.net
Adding new nodes
Hi All Could someone explain to me how the following is done - when new nodes are added how do we read existing data since the topology changes? How does Cassandra ensure that reads and writes are successful? Cheers Bill
Re: Adding new nodes
Data is moved to the new correct nodes. On Thu, Feb 4, 2010 at 10:52 PM, Bill Hastings bllhasti...@gmail.com wrote: Hi All Could someone explain to me how the following is done - when new nodes are added how do we read existing data since the topology changes? How does Cassandra ensure that reads and writes are successful? Cheers Bill
Re: Adding new nodes
Sorry I guess I was not clear enough. While the existing data is being moved do requests for reads go to the new nodes? If so what if that data has not yet migrated? There is no problem for writes. But how is the routing for the reads handled in this situation? Cheers Avinash On Thu, Feb 4, 2010 at 9:07 PM, Jonathan Ellis jbel...@gmail.com wrote: Data is moved to the new correct nodes. On Thu, Feb 4, 2010 at 10:52 PM, Bill Hastings bllhasti...@gmail.com wrote: Hi All Could someone explain to me how the following is done - when new nodes are added how do we read existing data since the topology changes? How does Cassandra ensure that reads and writes are successful? Cheers Bill
Re: Adding new nodes
Hi All I just had a conversation with one of the FB guys (Avinash) at FB and landed up signing off as him :). He wasn't quite sure about how this works in the OSS branch. Hence the question to the broader audience. The question is more about the change in topology and reads going to a machine before data migration. Cheers Bill On Thu, Feb 4, 2010 at 9:36 PM, Bill Hastings bllhasti...@gmail.com wrote: Sorry I guess I was not clear enough. While the existing data is being moved do requests for reads go to the new nodes? If so what if that data has not yet migrated? There is no problem for writes. But how is the routing for the reads handled in this situation? Cheers Avinash On Thu, Feb 4, 2010 at 9:07 PM, Jonathan Ellis jbel...@gmail.com wrote: Data is moved to the new correct nodes. On Thu, Feb 4, 2010 at 10:52 PM, Bill Hastings bllhasti...@gmail.com wrote: Hi All Could someone explain to me how the following is done - when new nodes are added how do we read existing data since the topology changes? How does Cassandra ensure that reads and writes are successful? Cheers Bill
Re: Adding new nodes
Hi All First off Bill I don't think I brainwashed you to an extent where you start signing off as me :). Don't do that on my check books. That's an interesting question and like I had said I am not too sure about this is handled in the current OSS version. Jonathan is your best bet for this response. Writes will be handled fine like we discussed. Cheers Avinash On Thu, Feb 4, 2010 at 9:46 PM, Bill Hastings bllhasti...@gmail.com wrote: -- Forwarded message -- From: Bill Hastings bllhasti...@gmail.com Date: Thu, Feb 4, 2010 at 9:41 PM Subject: Re: Adding new nodes To: cassandra-user@incubator.apache.org Hi All I just had a conversation with one of the FB guys (Avinash) at FB and landed up signing off as him :). He wasn't quite sure about how this works in the OSS branch. Hence the question to the broader audience. The question is more about the change in topology and reads going to a machine before data migration. Cheers Bill On Thu, Feb 4, 2010 at 9:36 PM, Bill Hastings bllhasti...@gmail.comwrote: Sorry I guess I was not clear enough. While the existing data is being moved do requests for reads go to the new nodes? If so what if that data has not yet migrated? There is no problem for writes. But how is the routing for the reads handled in this situation? Cheers Avinash On Thu, Feb 4, 2010 at 9:07 PM, Jonathan Ellis jbel...@gmail.com wrote: Data is moved to the new correct nodes. On Thu, Feb 4, 2010 at 10:52 PM, Bill Hastings bllhasti...@gmail.com wrote: Hi All Could someone explain to me how the following is done - when new nodes are added how do we read existing data since the topology changes? How does Cassandra ensure that reads and writes are successful? Cheers Bill -- Cheers Bill
AUTO: Hernan Badenes is out of the office (returning 02/07/2010)
I am out of the office until 02/07/2010. For any urgent matter, please contact Julian Ariel Cerruti/Argentina/IBM Note: This is an automated response to your message Adding new nodes sent on 5/2/10 1:52:47. This is the only notification you will receive while this person is away.
Re: Adding new nodes
if you are using the Right API's. read requests will not be sent to the bootstrapping nodes where as writes will be sent. Regards, /VJ On Thu, Feb 4, 2010 at 10:04 PM, Bill Hastings bllhasti...@gmail.comwrote: there will be reads that will fail in the interim. Am I way off here? Apologies if I am wrong and I will continue looking. I am very curious is knowing how this works.