Load Balancing and Highavailability Geographically using HAProxy

[xsysxpert@gmail.com]

Hello

Im new to NoSQL and i have sql thoughts yet , now for some reasons i need to migrate from mysql/mariadb to a nosql base on our programmers need .

In mariadb i use Galera for clustering and Haproxy for load balancing send requests based on geo or ... to db servers now after searching i findout maybe scylla is what i need
and want to use haproxy and i thinks its not a big deal if clustering does his job well  now i have master slave databases geographically in my current mariadb/galera zone but this case i need master-master at least 6 servers located in 3 continent ( i know that it contains risk if they have a lot latency but i have to try) so i need to know is clustering in scylladb works correct ? or like influxdb has some limitations before I start digging  about influxdb and find outs in clustering i can have only 3 servers . if limitation exist explain about them for me , and ofcourse i need a good documentation about scylladb clustering life cycle that i cant find it yet

[Tzach Livyatan]

On Sun, Jan 24, 2016 at 12:21 PM, <xsys...@gmail.com> wrote:

Hello

Hi 

 

Im new to NoSQL and i have sql thoughts yet , now for some reasons i need to migrate from mysql/mariadb to a nosql base on our programmers need .

In mariadb i use Galera for clustering and Haproxy for load balancing send requests based on geo or ... to db servers now after searching i findout maybe scylla is what i need
and want to use haproxy and i thinks its not a big deal if clustering does his job well  now i have master slave databases geographically in my current mariadb/galera zone but this case i need master-master at least 6 servers located in 3 continent ( i know that it contains risk if they have a lot latency but i have to try) so i need to know is clustering in scylladb works correct ? or like influxdb has some limitations before I start digging  about influxdb and find outs in clustering i can have only 3 servers . if limitation exist explain about them for me , and ofcourse i need a good documentation about scylladb clustering life cycle that i cant find it yet

ScyllaDB HA is based on the same proven mechanism of Apache Cassandra, including symmetrical nodes and tunable consistency.

In particular, each node in the cluster can act as a coordinator (proxy), so there is not need for additional HAProxy.  

Scylla supports multi region as Data Centers, where each DC can have different level of replication.  This deck[1] gives a good introduction to Cassandra HA, where slides #23 and #24 covers the Multi DC part. 

In short: Cassandra HA, in particular across regions (and continents) is well proven and very widely used. Scylla take the excellent HA of Cassandra, and completely rewrite each node for much better throughput and latency[2] 

Please let me know if you need more info, I will be happy to provide it.

If you want to take Scylla for a spin, you can start here

Regards

Tzach

ScyllaDB Product Manager

  

[1] http://www.slideshare.net/NaderGan/cassandra-101-26132602

[2] http://www.scylladb.com/technology/cassandra-vs-scylla-benchmark-cluster-1/

 

--
You received this message because you are subscribed to the Google Groups "ScyllaDB users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scylladb-user...@googlegroups.com.
To post to this group, send email to scyllad...@googlegroups.com.
Visit this group at https://groups.google.com/group/scylladb-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/scylladb-users/e13f7a70-3cbd-4525-aa8c-530b8b2fc9f3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[xsysxpert@gmail.com]

Thank you very much sir i will ask my questions after building the project and a quit working with that