I don't think such benchmark is necessary, it's really hard to say it's apple to apple case. And as you said, each system has it's pros and cons, considering your engine is based on Elasticsearch and ours on HBase, it will (probably) directing the discussion to compare them.
My question to you is what's the purpose you would like to do such comparison? With all above discussion, I think you already have information. And if you would like to propose ES is better than HBase, we are really welcome contribution to offer alternative storage option for users as same as we have being ask about Reddis. It doesn't matter whether a cat is black or white as long as it can catch mice. --by Xiaoping Deng. Thanks. Best Regards! --------------------- Luke Han On Thu, Nov 19, 2015 at 12:37 AM, Sarnath <[email protected]> wrote: > Hi Seshu, > I am not asking you guys help to benchmark another system. Bin said that > the test data was small and invalid for any reasonable comparison. So I am > merely asking for pointers to any public dataset that can be used. > Or if you could guys could tell me, what could be desirable properties of a > synthetic dataset, I would create that in an amazon cluster and benchmark. > Every system has its strength and weakness. And with big data, there are so > many ways to solve a same problem. Benchmarking is the best way to > understand. All of us can learn from such an exercise. > Best, > Sarnath >
