Tag

performance Archives - AWS Managed Services by Anchor

Making Magento Shine with Varnish – Part 1

By | General, Technical | One Comment

Developing for the web can be overwhelming – the stack of technologies involved has only grown over the years, whilst customers demand faster and more responsive websites. Performance is often an afterthought, partly because it can be tricky to define. New features are tangible and easily demonstrated, but it can be difficult to make a business case for performance during the development stage. Yet as studies by Akamai, Google and Amazon have shown, the success of e-commerce sites in particular is closely linked to how they perform. Magento is a popular e-commerce framework that offers a wealth of customisation through an extensible design, though this flexibility can easily result in slow, sluggish websites if you aren’t careful. But what does it even mean for a website to be slow or…

Read More

Kicking the tyres on OpenTSDB

By | Technical | No Comments

One of the big buzzwords in IT at the moment is “big data” – data in such large quantities that it’s not feasible to analyse using your traditional set of tools. Thankfully this isn’t a problem that Anchor has to deal with, but we almost wish we did. We collect a lot of data about the servers we manage, way more than most other hosting providers: on a typical server we monitor and track a couple dozen metrics to know how healthy it is and whether that’s changing over time. This is good, but it’d be great if we could easily store lots more data. What if we didn’t limit ourselves to keeping a year of data? What if we collected data every few seconds instead of every minute? Even…

Read More

Noop I/O scheduling with SSD storage

By | Technical | No Comments

Solid State Drives (SSDs) are not new in the server world, but they’re seen a somewhat limited takeup due to their high cost per gigabyte of storage. This has been changing as prices continue to drop, and we’re now at the point where SSDs represent a viable option for primary storage of high-value data that is read and written heavily. While SSDs are a drop-in replacement for traditional hard drives (HDDs) when it comes to servers, their behaviour and performance characteristics are fundamentally different. This means that the disk-access scheduling algorithms that we’ve developed over time, originally designed for rotational media, simply don’t work for SSDs. We recently observed some interesting behaviour as a result of changing disk access schedulers, which we thought we’d share with you here. The difference…

Read More

Turbocharging work queues for TestFlight

By | Technical | No Comments

We’ve recently been hard at work helping TestFlight scale up their operations. They do a lot of heavy lifting with redis, celery and mysql so most of the work centres around finding and eliminating bottlenecks. The biggest gains have come from optimising their use of mysql from the worker processes’ standpoint, resulting in a three-fold improvement in throughput. If you’re interested, have a look at TestFlight’s case study. It’s one part of the overall improvements that we’re pushing but it’s a great return for a small investment, which makes it worth noting. There’s also plenty more on the horizon for them, which we’re excited about. 🙂

Read More

Dedicated crypto accelerator cards? Please, that’s so last decade

By | Technical | No Comments

Today I’ve been looking over the legacy architecture for a new customer we have coming on board. I think it’s fair to say that they’re of a substantial size. One of the things that stands out to me is that they have five load-balancers (huh?) on the public-facing end, and then seven nginx frontends terminating SSL traffic and serving static content. Let’s ignore the load-balancers, I think they’re just some cloud-y appliance. The frontends are where it’s at. These are some pretty substantial VMs (a certain provider’s 2gb instance) running SSL all day and not much else. Their app doesn’t even run on the frontends! SSL crypto is very much the lifeblood of internet commerce, it’d come to a screeching halt without it these days. It’s just an unfortunate fact…

Read More

Your Magento store + Anchor = ?

By | Company News | 2 Comments

A little bit of horn-blowing, the correct answer is of course “a winning combination”. 🙂 We often find ourselves bothered by PHP instead of being hot-and-bothered, but Magento is a pretty well-engineered app. It’s got solid documentation (a godsend), and while it’s very resource intensive if you’re a $5-a-month hosting customer, it’s clear they’ve given a lot of thought to scalability for running a serious online shop. Scalability? Yes please! If you’re interested in that sort of thing, we recently published a little case study about our friends at Games Paradise, and how we helped them gear up for the Christmas season. To give you an idea of what we’re talking about (in a pretty little graph): Games Paradise isn’t the only one, mind you. We’ve also helped other customers…

Read More

The Myth of Infinite Cloud Scalability

By | Company News, Technical | No Comments

Recently, I read an article from a fairly prominent “cloud computing” vendor, which contained a line that basically said “Let the cloud worry about your scalability and performance problems”. I nearly snorted my late-mid-morning can of mother out my nose when I read it. Here’s why. “Let the cloud worry about your scalability” is nothing more than a thinly disguised version of “just throw more hardware at it”. This is a “solution” beloved of salespeople everywhere, because it’s plausible, real easy to say, and makes a whole pile more money for the company providing the hardware. However, while it can be an appropriate solution in the right circumstances, and with appropriate evidence of its effectiveness in those particular circumstances, it usually isn’t the only option, it often isn’t the best…

Read More

Reduce Linux VPS/VM guest memory usage

By | Technical | 3 Comments

Reducing the memory usage in your VPS/VM can be a great way to free up some resources to handle more requests, users or some other metric of win. By default at Anchor we provision our Red Hat & Cent OS VPS servers with a trim memory usage profile by disabling a lot of unneeded services at install time. We do this by using Trogdor (our hardware/software burninator) and Puppet. So just what services do we disable, if they exist on the new VPS? gpm netfs pcmcia sgi_fam yum-updatesd pcscd rhnsd xfs hald hcid or sdpd hpiod or hpssd.py dbus-daemon cupsd You can also reap performance gains by changing how you serve content. For example you can use a cut down high performance web server (nginx or lighttpd) to serve all…

Read More

nsscache and LDAP reliability

By | Technical | No Comments

Any company with multiple servers in their authentication domain will know of LDAP. Sadly on the Linux platform, OpenLDAP (although arguably the most widely used and well known of the few LDAP servers available) is still not particularly reliable, especially when it comes to replication. The overheads involved in querying even a local OpenLDAP server are much higher than, say, the plaintext files such as /etc/passwd. Enter nsscache. Created by two boffins at Google (one of whom graduated from Anchor Systems), nsscache gives the reliability and speed of plaintext files (or BDB if you desire) and the scalability of OpenLDAP. Anchor recently started using it and we are confident it will dramatically boost the reliability and lookup speed of all of our LDAP systems. In terms of performance, we are…

Read More