<antirez>

Yesterday Amplitude published an article about scaling analytics, in the context of using the Set data type. The blog post is here: https://amplitude.com/blog/2015/08/25/scaling-analytics-at-amplitude/

On Hacker News people asked why not using Redis instead: https://news.ycombinator.com/item?id=10118413 

Amplitude developers have their set of reasons for not using Redis, and in general if you have a very specific problem and want to scale it in the best possible way, it makes sense to implement your vertical solution. I’m not adverse to reinventing the wheel, you want your very specific wheel sometimes, that a general purpose system may not be able to provide. Moreover creating your solution gives you control on what you did, boosts your creativity and your confidence in what you, as a developer can do, makes you able to debug whatever bug may arise in the future without external help.

I consider myself very lucky for contributing to the open source. For me OSS software is not just a license: it means transparency in the development process, choices that are only taken in order to improve software from the point of view of the users, documentation that attempts to cover everything, and simple, understandable systems. The Redis community had the privilege of finding in Pivotal, and VMware before, a company that thinks at open source in the same way as we, the community of developers, think of it.

Nor subjects, for what matters. Everybody will tell you to don't add a dot at the end of the first line of a commit message. I followed the advice for some time, but I'll stop today, because I don't believe commit messages are titles or subjects. They are synopsis of the meaning of the change operated by the commit, so they are small sentences. The sentence can be later augmented with more details in the next lines of the commit message, however many times there is *no* body, there is just the first line. How many emails or articles you see with just the subject or the title? Very little, I guess. So for me it is like:

I’m back from Paris, DotScale 2015 was a very interesting conference. Before leaving I was working on Sentinel in the context of the unstable branch: the work was mainly about connection sharing. In short, it is the ability of a few Sentinels to scale, monitoring many masters. Before to leave, and now that I’m back, I tried to “secure” a set of features that will be the basis for Redis 3.2. In the next weeks I’ll be focusing developing these features, so I thought it’s worth to share the list with you ASAP.

EDIT: In case you missed it, Disque source code is now available at http://github.com/antirez/disque

It is a few months that I spend ~ 15-20% of my time, mostly hours stolen to nights and weekends, working to a new system. It’s a message broker and it’s called Disque. I’ve an implementation of 80% of what was in the original specification, but still I don’t feel like it’s ready to be released. Since I can’t ship, I’ll at least blog… so that’s the story of how it started and a few details about what it is.

I’m back home, after a non easy trip, since to travel from San Francisco to Sicily is kinda NP complete: there are no solutions involving less than three flights. However it was definitely worth it, because the Redis Conference 2015 was very good, SF was wonderful as usually and I was able to meet with many interesting people. Here I’ll limit myself to writing a short account of the conference, but the trip was also an incredible experience because I discovered old and new friends, that are not just smart programmers, but also people I could imagine being my friends here in Sicily. I never felt alone while I was 10k kilometers away from my home.

Today Redis is six years old. This is an incredible accomplishment for me, because in the past I switched to the next thing much faster. There are things that lasted six years in my past, but not like Redis, where after so much time, I still focus most of my everyday energies into.

How did I stopped doing new things to focus into an unique effort, drastically monopolizing my professional life? It was a too big sacrifice to do, for an human being with a limited life span. Fortunately I simply never did this, I never stopped doing new things.

Redis speed could be one selling point for new users, so following the trend of comparative “advertising” it should be logical to have a few comparisons at Redis.io. However there are two problems with this. One is of goals: I don’t want to convince developers to adopt Redis, we just do our best in order to provide a suitable product, and we are happy if people can get work done with it, that’s where my marketing wishes end. There is more: it is almost always impossible to compare different systems in a fair way.

Today I was testing Redis latency using m3.medium EC2 instances. I was able to replicate the usual latency spikes during BGSAVE, when the process forks, and the child starts saving the dataset on disk. However something was not as expected. The spike did not happened because of disk I/O, nor during the fork() call itself.

The test was performed with a 1GB of data in memory, with 150k writes per second originating from a different EC2 instance, targeting 5 million keys (evenly distributed). The pipeline was set to 4 commands. This translates to the following command line of redis-benchmark:

One interesting thing about the Stripe blog post about Redis is that they included latency graphs obtained during their tests. In order to persist on disk Redis requires to call the fork() system call. Usually forking using physical servers, and most hypervisors, is fast even with big processes. However Xen is slow to fork, so with certain EC2 instance types (and other virtual servers providers as well), it is possible to have serious latency spikes every time the parent process forks in order to persist on disk. The Stripe graph is pretty clear in this regard.