ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster

>The Scylla design, right, is based on a modern shared-nothing approach. Scylla runs multiple engines, one per core, each with its own memory, CPU and multi-queue NIC. We can easily reach 1 million CQL operations on a single commodity server. In addition, Scylla targets consistent low latency, under 1 ms, for inserts, deletes, and reads.

Interesting. From: http://www.scylladb.com/technology/architecture/

Re: ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster

#5

Post by domlebo70 » Wed, Sep 23, 2015, 2:08 AM UTC

Literally zero mention that in the event of a network partition they will just drop messages on the floor. This is fine as a cache, perhaps replacing Redis... but as a Cassandra replacement this is pretty scary

Re: ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster

#6

Post by whalesalad » Wed, Sep 23, 2015, 3:12 AM UTC

nattaylor wrote: post #4

>The Scylla design, right, is based on a modern shared-nothing approach. Scylla runs multiple engines, one per core, each with its own memory, CPU and multi-queue NIC. We can easily reach 1 million CQL operations on a single commodity server. In addition, Scylla targets consistent low latency, under 1 ms, for inserts, deletes, and reads. Interesting. From: http://www.scylladb.com/technology/architecture/

So on virtualized hardware, namely AWS, I'm sure the benchmarks won't be so magnificent. Needing a dedicated nic per core is a big deal unless you're at a pretty large scale.

Re: ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster

#7

Post by nnx » Wed, Sep 23, 2015, 3:12 AM UTC

domlebo70 wrote: post #5

Literally zero mention that in the event of a network partition they will just drop messages on the floor. This is fine as a cache, perhaps replacing Redis... but as a Cassandra replacement this is pretty scary

Source?

Re: ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster

#8

Post by vitalyd » Wed, Sep 23, 2015, 3:14 AM UTC

domlebo70 wrote: post #5

Literally zero mention that in the event of a network partition they will just drop messages on the floor. This is fine as a cache, perhaps replacing Redis... but as a Cassandra replacement this is pretty scary

Where did you get this from? I hope that's not a conclusion from the benchmark doing single server load testing.

Re: ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster

#9

Post by vitalyd » Wed, Sep 23, 2015, 3:16 AM UTC

whalesalad wrote: post #6

nattaylor wrote: post #4

>The Scylla design, right, is based on a modern shared-nothing approach. Scylla runs multiple engines, one per core, each with its own memory, CPU and multi-queue NIC. We can easily reach 1 million CQL operations on a single commodity server. In addition, Scylla targets consistent low latency, under 1 ms, for inserts, deletes, and reads. Interesting. From: http://www.scylladb.com/technology/architecture/

So on virtualized hardware, namely AWS, I'm sure the benchmarks won't be so magnificent. Needing a dedicated nic per core is a big deal unless you're at a pretty large scale.

Not dedicated NIC per core, but multi-queue NIC having its queues serviced by dedicated cores.

Re: ScyllaDB: Drop-in replacement for Cassandra that claims to be 10x faster

#10

Post by dschiptsov » Wed, Sep 23, 2015, 4:13 AM UTC

Finally, back to sanity of great old-school products, like Informix, by dropping Java (the whole scam) for C++14 and by paying attention to details of an underlying OS (again).

Same trend, by the way, is in Android development.