benchmarking - What is the maxium replication rate of Couchbase XDCR -


we using couchbase data caching , there talk of doing cross-data center replication it. however, need 1000 documents replicated multiple locations every second. documents between 2 , 64k each.

is there out there xdcr experience can tell me whether feasible, or if have use other means replicate data @ speed. "benchmark" in documentation @ couchbase implies rate of xdcr 100tps. (149 ms replicate 11 documents.)

the replication rate of xdcr limited network bandwidth , latency first, cpu , disk io. assuming have enough bandwidth between datacenters , clusters provisioned properly, couchbase replicate hundreds of thousands of documents per second, or more. it's pretty simple experiment run, set xdcr between 2 singles node clusters , use 1 of load generator tools come couchbase create traffic. (cbworkloadgen in couchbase bin folder or cbc-pillowfight comes libcouchbase.)

there several config settings can play optimize throughput, such increasing batch size, changing optimistic replication threshold, etc.


Comments

Popular posts from this blog

OpenCV OpenCL: Convert Mat to Bitmap in JNI Layer for Android -

python - How to remove the Xframe Options header in django? -

android - org.xmlpull.v1.XmlPullParserException: expected: START_TAG {http://schemas.xmlsoap.org/soap/envelope/}Envelope -