1. 16
  1.  

  2. 3

    This OnGress company keeps repeating about their “50 pages whitepaper” but in order to get that you need to fill in a form that requires an email address and phone number. https://info.enterprisedb.com/Performance-Benchmarks-PostgreSQL-vs-MongoDB.html

    When you benchmark, you have to make sure you eliminate external overhead. This OnGress use a driver didn’t support connection pooling…They said “I disagree, because this means that this is the performance that Lua users get. “…So yeah, they are benchmarkching Lua driver after all.

    1. 2

      When you benchmark, you have to make sure you eliminate external overhead. This OnGress use a driver didn’t support connection pooling…They said “I disagree, because this means that this is the performance that Lua users get. “…So yeah, they are benchmarkching Lua driver after all.

      This is the full paragraph.

      Sysbench is written in Lua. Because of this, MongoDB concludes that “any reasonable tester would have looked for an alternative benchmark”. I disagree, because this means that this is the performance that Lua users get. Moreover, it tells the world how much MongoDB actually cares about Lua users, specially if this driver (which was created by MongoDB) remains unmaintained, and no other official Lua driver has been created.

      It seems like they used the official MongoDB driver, which seems totally reasonable. If Mongo’s own driver sucks and skews the test results then I agree that it isn’t necessarily fair to conclude the Mongo database itself is slow, but I do think it’s reasonable to conclude that Mongo doesn’t care much about Lua users.

      Having said that, Mongo is slow IMO. My employer is a paying customer on their hosted platform. I’m always surprised how slow it is doing anything that isn’t a lookup by ObjectID or when an index already exists.