Since our last update, version 2.3 shipped, including some huge efficiency wins, and a whole bunch of bug fixes.
Seq 2.4 is also just around the corner, and continues the theme of making sure Seq can keep up with our customers.
Version 2.4.1-pre is the first build of Seq to make use of real-time compaction (on Windows 8.1 and Windows Server 2012 R2), eliminating the heavy disk I/O spikes and space requirements sometimes incurred by periodic compaction.
Further optimization of filter execution also brings more performance gains in 2.4, especially for the complex filters associated with signals.
The current stable and preview releases are available on the download page of the Seq website.
Roadmap
With Seq 2 now feeling very well-rounded, it's time to start sharing some of our plans for what's coming down the line.
The first feature scheduled for Seq's next major version is aggregate queries, like min, max, sum, count, mean and percentile. Being able to roll up values over a time intervals is fundamental to a whole range of scenarios.
To enable this in Seq we're extending the query syntax to borrow some familiar SQL constructs:
select mean(OrderSize) group by StoreId, time(1h)
In a stream of order processing events, this query will return the average order size, per store, across one hour intervals.
As you can imagine, aggregate query support opens up a lot of exciting analysis opportunities -- we can't wait to land this change. A lot of code is written, but there's still plenty to go. More details will be posted here once we have a working preview for everyone to try.
Documentation updates
The Seq documentation continues to expand. If there are any topics you'd like to see added to the documentation, please reply!
Recently added content includes:
- Azure Installation Guide - a handy reference whether it is your first time setting up an Azure virtual machine, or if you just need a checklist. to avoid missing the pesky Windows Firewall configuration step!
- POSTing Raw Events - send events to Seq from practically anywhere using HTTP and a simple JSON format.