Saving $500k per month buying your own hardware: cloud vs co-location

By David Mytton,
CEO & Founder of Server Density.

Published on the 23rd November, 2016.

Editor’s note: This is an updated version of an article originally published on GigaOm on 07/12/2013.

A few weeks ago we compared cloud instances against dedicated servers. We also explored various scenarios where it can be significantly cheaper to use dedicated servers instead of cloud services.

But that’s not the end of it. Since you are still paying on a monthly basis then if you project the costs out over 1 to 3 years, you end up paying much more than it would have cost to outright purchase the hardware. This is where buying and co-locating your own hardware becomes a more attractive option.

Putting the numbers down: cloud vs co-location

Let’s consider the case of a high throughput database hosted on suitable machines on cloud and dedicated servers and on a purchased/co-located server. For dedicated instances, Amazon has a separate fee structure and on Rackspace you effectively have to get their largest instance type.

So, calculating those costs out for our database instance on an annual basis would look like this:

Amazon EC2 c3.4xlarge dedicated heavy utilization reserved
Pricing for 1-year term
$4,785 upfront cost
$0.546 effective hourly cost
$2 per hour, per region additional cost
$4,785 + ($0.546 + $2.00) * 24 * 365 = $27,087.96

Rackspace OnMetal I/O
Pricing for 1-year term
$2.46575 hourly cost
$0.06849 additional hourly cost for managed infrastructure
Total Hourly Cost: $2.53424
$2.53424 * 24 * 365 = $22,199.94

Softlayer

Given the annual cost of these instances, it makes sense to consider dedicated hardware where you rent the resources and the provider is responsible for upkeep. Here, at Server Density, we use Softlayer, now owned by IBM, and have dedicated hardware for our database nodes. IBM is becoming very competitive with Amazon and Rackspace so let’s add a similarly spec’d dedicated server from SoftLayer, at list prices. To match a similar spec we can choose the Monthly Bare Metal Dual Processor (Xeon E5-2620 – 2.0Ghz, 32GB RAM, 500GB storage). This bears a monthly cost of $491 or $5,892/year.

Dedicated servers summary

Rackspace Cloud Amazon EC2 Softlayer Dedicated
$22,199.54 $27,087.96 $5,892

Let’s also assume purchase and colocation of a Dell PowerEdge R430 (two 8-core processors, 32GB RAM, 1TB SATA disk drive).

The R430 one-time list price is $3,774.45 – some 36% off the price of the SoftLayer server at $5,892/year. Of course there might be some more usage expenses such as power and bandwidth, depending on where you choose to colocate your server. Power usage in particular is difficult to calculate because you’d need to stress test the server, figure out the maximum draw and run real workloads to see what your normal usage is.

Running our own hardware

We have experimented with running our own hardware in London. In order to draw some conclusions we decided to use our 1U Dell server that has specs very similar to Dell R430 above. With everyday usage, our server’s power needs range close to 0.6A. For best results we stress tested it with everything maxed, for a total of 1.2A.

Hosting this with the ISP who supplies our office works out at $140/month or $1,680/year. This makes the total annual cost figures look as follows:

Rackspace Cloud Amazon EC2 Softlayer Dedicated Co-location
$22,199.54 $27,087.96 $5,892 $5,454.45/year 1, then $1,680/year

With Rackspace, Amazon and SoftLayer you’d have to pay the above price every year. With co-location, on the other hand, after the first year the annual cost drops to $1,680 because you already own the hardware. What’s more, the hardware can also be considered an asset yielding tax benefits.

Large scale implementation

While we were still experimenting on a small scale, I spoke to Mark Schliemann, who back then was VP of Technical Operations at Moz.com. They’d been running a hybrid environment and they had recently moved the majority of their environment off AWS and into a colo facility with Nimbix. Still, they kept using AWS for processing batch jobs (the perfect use case for elastic cloud resources).

Moz worked on detailed cost comparisons to factor in the cost of the hardware leases (routers, switches, firewalls, load balancers, SAN/NAS storage & VPN), virtualization platforms, misc software, monitoring software/services, connectivity/bandwidth, vendor support, colo, and even travel costs. Using this to calculate their per server costs meant that on AWS they would spend $3,200/m vs. $668/m with their own hardware. Their calculations resulted in costs of $8,096 vs. $38,400 at AWS, projecting out 1 year.

Optimizing utilization is much more difficult on the cloud because of the fixed instance sizes. Moz found they were much more efficient running their own systems virtualized because they could create the exact instance sizes they needed. Cloud providers often increase CPU allocation alongside memory whereas most use cases tend to rely on either one or the other. Running your own environment allows you to optimize this balance, and this was one of the key ways Moz improved their utilization metrics. This has helped them become more efficient with their spending.

Here is what Mark told me: “Right now we are able to demonstrate that our colo is about 1/5th the cost of Amazon, but with RAM upgrades to our servers to increase capacity we are confident we can drive this down to something closer to 1/7th the cost of Amazon.”

Co-location has its benefits, once you’re established

Co-location looks like a winner but there are some important caveats:

  • First and foremost, you need in-house expertise because you need to build and rack your own equipment and design the network. Networking hardware can be expensive, and if things go wrong your team needs to have the capacity and skills to resolve any problems. This could involve support contracts with vendors and/or training your own staff. However, it does not usually require hiring new people because the same team that deals with cloud architecture, redundancy, failover, APIs, programming, etc, can also work on the ops side of things running your own environment.
  • The data centers chosen have to be easily accessible 24/7 because you may need to visit at unusual times. This means having people on-call and available to travel, or paying remote hands at the data center high hourly fees to fix things.
  • You have to purchase the equipment upfront which means large capital outlay (although this can be mitigated by leasing.)

So what does this mean for the cloud? On a pure cost basis, buying your own hardware and colocating is significantly cheaper. Many will say that the real cost is hidden in staffing requirements but that’s not the case because you still need a technical team to build your cloud infrastructure.

At a basic level, compute and storage are commodities. The way the cloud providers differentiate is with supporting services. Amazon has been able to iterate very quickly on innovative features, offering a range of supporting products like DNS, mail, queuing, databases, auto scaling and the like. Rackspace was slower to do this but has already started to offer similar features.

Flexibility of cloud needs to be highlighted again too. Once you buy hardware, you’re stuck with it for the long term, but the point of the example above was that you had a known workload.

Considering the hybrid model

Perhaps a hybrid model makes sense, then? This is where I believe a good middle ground is. I know I saw Moz making good use of such a model. You can service your known workloads with dedicated servers and then connect to the public cloud when you need extra flexibility. Data centers like Equinix offer Direct Connect services into the big cloud providers for this very reason, and SoftLayer offers its own public cloud to go alongside dedicated instances. Rackspace is placing bets in all camps with public cloud, traditional managed hosting, a hybrid of the two, and support services for OpenStack.

And when should you consider switching? Nnamdi Orakwue, Dell VP of Cloud until late 2015, said companies often start looking at alternatives when their monthly AWS bill hits $50,000 but is even this too high?

Free eBook: The 9 Ingredients of Scale

From two students with pocket money, to 20 engineers and 80,000 servers on the books, our eBook is a detailed account of how we scaled a world-class DevOps team from the ground up. Download our definitive guide to scaling DevOps and how to get started on your journey.

Help us speak your language. What is your primary tech stack?

What infrastructure do you currently work with?

Articles you care about. Delivered.

Help us speak your language. What is your primary tech stack?

Maybe another time