Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
1–10 of 38 posts
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#2> Since the Redis cluster version is still in development, we
implemented our own YCSB client using the sharding capabilities of the Java Jedis library.
I just glanced over this, but at this point they might have very well benchmarked driver performance. I also don't see why somebody would compare Cassandra against MySQL and against Redis.
While they all persist application data, their feature sets are so completely different that I don't see for which task people would have to decide between them.
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#3It's refreshing to see a benchmark that is somewhat thorough and doesn't even make sweeping statements like "X is faster than Y".
Everyone who's ever written a hello world benchmark? Learn from this.
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#4> Since the Redis cluster version is still in development, we implemented our own YCSB client using the sharding capabilities of the Java Jedis library. I just glanced over this, but at this point they might have very well benchmarked driver performance. I also don't see why somebody would compare Cassandra against MySQL and against Redis. While they all persist application data, their feature sets are so completely…
I think they wanted to discriminate between them on tasks where someone might reasonably decide for either of them.
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#5It's refreshing to see a benchmark that is somewhat thorough and doesn't even make sweeping statements like "X is faster than Y". Everyone who's ever written a hello world benchmark? Learn from this.
Well they did say that Cassandra is better than all the others - especially when compared to HBase - on nearly all measurements (except for high write scenario latencies.)
This makes the decision of Facebook to go with HBase for their new messaging platform back in 2010 all the more strange. Though that was two years ago so things might have changed in Cassandra's favor since then.
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#6> Since the Redis cluster version is still in development, we implemented our own YCSB client using the sharding capabilities of the Java Jedis library. I just glanced over this, but at this point they might have very well benchmarked driver performance. I also don't see why somebody would compare Cassandra against MySQL and against Redis. While they all persist application data, their feature sets are so completely…
Even though Redis is moving from being a Memcached competitor to being more of a Cassandra competitor and MySQL is used by many of the world's largest websites (often as a glorified key-value store) they are odd choices.
Would anyone really use Redis for 'big data' ?
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#7> Since the Redis cluster version is still in development, we implemented our own YCSB client using the sharding capabilities of the Java Jedis library. I just glanced over this, but at this point they might have very well benchmarked driver performance. I also don't see why somebody would compare Cassandra against MySQL and against Redis. While they all persist application data, their feature sets are so completely…
Even though Redis is moving from being a Memcached competitor to being more of a Cassandra competitor and MySQL is used by many of the world's largest websites (often as a glorified key-value store) they are odd choices. Would anyone really use Redis for 'big data' ?
None of them do auto-sharding and master-master replication.
This is the domain dynamo/bigtable-like systems (elasticsearch, riak, cassandra, hbase, voldemort)
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#8> Since the Redis cluster version is still in development, we implemented our own YCSB client using the sharding capabilities of the Java Jedis library. I just glanced over this, but at this point they might have very well benchmarked driver performance. I also don't see why somebody would compare Cassandra against MySQL and against Redis. While they all persist application data, their feature sets are so completely…
I think they wanted to discriminate between them on tasks where someone might reasonably decide for either of them.
I haven't seen those usecases yet.
Either you want to scale over multiple systems or you don't. Either you have more data than RAM or you don't.
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#9This is like publishing the comparative performances of a Nissan Altima, a Ford F-150, a Honda Odyssey, a Smart Fortwo, a Kawasaki Ninja, and a Gillig Advantage.
Sure, it's possible to determine which one has the best land speed, range, and fuel economy, but you're also sure as heck not going to use a Gillig Advantage to do an F-150's job or vice versa.
Re: Benchmarks of Cassandra, HBase, VoltDB, MySql, Voldemort and Redis
#10With these sorts of academic benchmarks I frequently find myself questioning what exactly got tested. The researchers obviously know what they're doing, but that doesn't make them experts on all of the technologies under test. Their description of their HBase configuration in particular leaves me skeptical of how closely it reflects a well-tuned production setup. I think publishing the actual configuration files they used for each technology would be a big help.