Programming

Processing Pipelines Series - Reactive Extensions (Rx.NET)

Whereas TPL Dataflow is all about passing messages between blocks, Reactive Extensions is about sequences. With sequences we can create projections, transformations and filters. We can combine multiple sequences into a single one. It is a very flexible and powerful paradigm but with such power comes extra complexity. I find TPL Dataflow easier to reason about due to its simple model. Reactive Extensions can get pretty complex and is not always intuitive, but you can create some elegant solutions with it. It will require some investment in time and tinkering to get a reasonable understanding of it.

Processing Pipelines Series - TPL Dataflow - Alternate Scenario

In the last post we built a TPL Dataflow pipeline based on the scenario from our first post in the series. Today we'll build another pipeline very similar to the first but with different requirements around latency and data loss.

In the first scenario we could not slow down the producer as slowing it down would cause data loss (it read from a bus that would not wait if you weren't there to consume the data). We also cared a lot about ensuring the first stage kept up with the producer and successfully wrote every message to disk. The rest was best effort, and we performed load-shedding so as not to slow down the producer.

Processing Pipelines Series - Concepts

In this series we'll look at few different .NET technologies we can use to process streams of data in processing pipelines and directed acyclic graphs (DAGs). This is not about distributed data platforms for big data but real-time processing and computation running on a single machine. We'll take a single scenario and build it out multiple times, each with a different technology. Each application will be built as a console application with .NET Core.

DSL Parser - Sample Code

I while back I wrote a blog series about DSLs, grammars, tokenizers, parsers and a SQL generator. The idea was that you could write a DSL query to mine your error log data and the code would generate SQL. The series can be found here: http://jack-vanlightly.com/blog/2016/2/3/how-to-create-a-query-language-dsl. 

The tokenizer while simple was very inefficient, so I wrote a better one, you can find that here: http://jack-vanlightly.com/blog/2016/2/24/a-more-efficient-regex-tokenizer

I have just published working code based on this series and the better Regex tokenizer on Github here: https://github.com/Vanlightly/DslParser.

How to Kill a Keep Alive with a Weak Reference (C#)

Taskling.NET uses a keep alive or heartbeat to signal that it is still running. This is useful because when running batch jobs in unstable hosts like IIS the process can be killed off with a ThreadAbortException and the job isn't always able to log it's demise. With a keep alive we know that the job really died if a few minutes pass without a keep alive and the status of the job is "In Progress".

But one problem is how do you reliably kill a keep alive?

How Row Locking Makes Taskling Concurrency Controls Possible

Your Taskling jobs can be configured with concurrency limits and those jobs will never have more than the configured number of executions of that job running at any time.

Some batch and micro-batch jobs need to be singletons, there to be only one execution running at any point in time. This may be to avoid data consistency issues when persisting results or because only a single session can be opened to a third party service etc. Other batch processes need more than one execution running at the same time in order to cope with the data volume but have a concurrency limit in order to not overwhelm downstream systems or third party services.