Geeks With Blogs

News This is the *old* blog. The new one is at
Elton Stoneman
This is the *old* blog. The new one is at
The concurrency shoot-out: SQL Server versus MongoDB (part 3)
In part 1 and part 2 we looked at running concurrent workflows in SQL Server and MongoDB with different configurations of context, isolation and journaling. MongoDB was by far the better performer, in all except Variant 4, where the client waited for the server to commit changes to the journal before acknowledging the response, at the default journal commit window of 100ms. In a production system, we’d be looking at using variant 2 or variant 5 of Mongo. If we don’t need ultra-reliability, then we ......

Posted On Sunday, January 5, 2014 9:29 PM

The concurrency shoot-out: SQL Server versus MongoDB (part 2)
Following up on The concurrency shoot-out: SQL Server versus MongoDB (part 1), I extended my performance tests with some variations to look at the options for performance and reliability, and see how they affected the results. Variant 1: reuse the database context In the first version of my performance tests, I used a new database context for each part of the workflow (instantiating a new EntityContext in EF, or a new MongoCollection in Mongo for parts 2, 3 and 4 of the workflow). Realistically, ......

Posted On Wednesday, December 18, 2013 4:24 PM

The concurrency shoot-out: SQL Server versus MongoDB (part 1)
One of the biggest performance issues with the solution we’re building at the moment has been contention in SQL Server. We have NServiceBus processing workflows and we’re distributing the load across multiple queues and multiple host processes, each with multiple threads. At high load we have something like 200 threads running concurrently, all trying to read and write data in the same tables. For each step in the workflow we record an event against the main entity, which gives us an audit trail ......

Posted On Friday, December 13, 2013 10:10 PM

Running a Mongo Replica Set on Azure VM Roles
Setting up a MongoDB Replica Set with a bunch of Azure VMs is straightforward stuff. Here’s a step-by-step which gets you from 0 to fully-redundant 3-node document database in about 30 minutes (most of which will be spent waiting for VMs to fire up). First, create yourself 3 VM roles, which is the minimum number of nodes you need for high availability. You can use any OS that Mongo supports. This guide uses Windows but the only difference will be the mechanism for starting the Mongo service when ......

Posted On Thursday, October 10, 2013 12:27 PM

Copyright © Elton Stoneman | Powered by: