Click here to LEARN more.

Nov 14, 2017

Choosing a Public Cloud: Avoiding Noisy Neighbors

Paul Painter, Director, Solutions Engineering

I frequently hear people asking how to choose from among so many potential cloud options. The answer, I’m afraid, is “It depends.” Some cloud providers may over-provision their resources, causing your traffic to slow down. This drop in service is often called the “noisy neighbor” syndrome, but how can you avoid this?

Location, Location, Location

Just like choosing a house, it’s important to get a comfort level with the neighborhood. To evaluate cloud provider environments, you need to ask a few questions:

1.) What is the underlying hardware in the cloud environment?

For performance issues, you first want to confirm that the processor is equal or greater in horsepower than your computing needs. It is also important to know if hyper-threading is activated on the processor. For example, our AgileCLOUD is built using Intel e5-2650v3 processors, which have 10 cores running at 3.0 GHz per core and hyper-threading enabled, effectively doubling the processing power.

2.) What is the vCPU ratio?

In a virtual cloud environment, the hypervisor will divide the CPU cores into smaller mini-CPUs or virtual CPUs (vCPUs). Many providers will oversubscribe the available vCPUs; in other words, they will assign more vCPUs than physically available. This assumes that running virtual servers requires all the CPU cycles assigned. Cloud environments that oversubscribe are often cheaper and are ideal for workloads that are typically idle most of the time.

When a guest operating system is installed on the hypervisor, the guest instance is assigned a pool of vCPUs, as well as virtual RAM (vRAM) and disk storage to be used. The hypervisor manages the time a vCPU has access to the physical CPU in a round-robin manner, thus creating vCPU queues.

3.) What are the underlying disk configuration and network speeds?

The local disk type (SSD or HDD), number of disks and the RAID configuration will affect the performance of an I/O intensive application. AgileCLOUD uses SSD disks with RAID10, providing the fastest local disk performance possible.

Hypervisors and the Noisy Neighbor Syndrome

Just like large lots make for quiet neighborhoods, a low physical to virtual ratio makes for good computing. A cloud with a low ratio will have more access to the physical hardware, like a house with a large yard keeps noisy neighbors’ music from disturbing your dinner. A higher ratio of resources means you have less access to the actual physical hardware, which would be like your townhouse neighbor’s rock band practicing next door.

Understanding a cloud provider’s performance is more complicated than vCPU price performance.

  • Compare the processor speeds: some cloud providers have different generations on hardware with different clock speeds (and cores).
  • What is the oversubscription ration?
  • Available Disk I/O, both read/write capacity of the drive and network access if attaching to a block storage device are critical to server performance.
  • Finally, understand the physical network capacity of the server to move both LAN/WAN data and any storage access.

INAP Provides You Options

Our AgileCLOUD for example, comes in two flavors and, depending on your workload, one may be better than the other. Our Series A is specifically suited for web, application and light I/O workloads. Our Series B is better suited for applications that have higher CPU and memory demands.

Answering which of these two options is better for you depends on the workload you have, and the answer may even be both!  Let me break down the differences for you. The obvious differentiator between the two options is the hypervisor on the AgileCLOUD lineup.

To keep it simple, our AgileCLOUD offers two series of cloud compute (see table below).

Our A Series, good for small databases, websites and content management systems that require moderate CPU utilization has a 3:1 vCPU to CPU ratio.

Our B Series, better for medium databases, complex websites and scheduled batch processing tasks requiring heavy memory and CPU utilization has a one to one (1:1) ratio.

In other words, we don’t overprovision on the B series, so there is no chance of CPU contention and little chance of that noisy neighbor.

Armed with this knowledge, we believe that you will be in a better position to understand the vast array of available cloud options. If you are interested in learning more about cloud solutions that can fit into your unique cloud strategy, contact us today to speak with one of our cloud professionals or deploy your instances right away in our cloud portal.

Explore HorizonIQ
Bare Metal


About Author

Paul Painter

Director, Solutions Engineering

Read More
Feb 26, 2015

Bare Metal vs. Hypervisor: The Evolution of Dedicated Servers

Paul Painter, Director, Solutions Engineering

With the “cloudification” of IT, new terms have appeared across the hosting industry. Some of the most popular terms are bare metal and hypervisor. But what are these? Let’s clarify the bare metal first. To answer simply, a bare metal server is your traditional dedicated server with a hip new name for the cloud generation! Let’s have a look at what a dedicated/bare metal server is.

What is Bare Metal?

Bare metal is a single tenant server. This means only you are taking the resources of the server. The server belongs to you and you only. Compared to the cloud model where multiple users (multi-tenancy) reside on the same physical server, the bare metal server only has one customer on the server.

Single tenancy allows you to avoid the noisy neighbor effect, which is described as a user impacting the performance and stability of other users within the same server. With bare metal, since you are the sole user, you will not witness the noisy neighbor effect.

From a financial perspective, the bare metal server is typically billed per month. This means no surprise on your bill at the end of the month. However, with the “cloudification” of hosting, bare metal is also available in hourly billing, so keep in mind that costs are variable when selecting an hourly usage plan.

Bare metal supports multiple types of operating systems on top of it, including hypervisors. This brings us to our next point—the difference between bare metal and hypervisors.

What is a Hypervisor?

How does hypervisor differ from bare metal? A hypervisor is an operating system that can create virtual machines (VM) within a bare metal server.

A traditional bare metal server: The operating system (CentOS, Debian, Redhat, SUSE, Ubuntu, Windows Server, etc.) is installed directly on the server, and applications are running natively in the operating system.

A bare metal server installed with a hypervisor provides the user with a management suite to create virtual machines on the server. The hypervisor should not run applications natively; rather, its purpose is to virtualize your workloads into separate virtual machines to gain the flexibility and reliability of virtualization.

Multiple hypervisors exist on the market. We have reached a point with virtualization technology where the ecosystem is very mature and all options are very similar. So the choice of a hypervisor relies mainly on the following points: your current familiarity with a vendor, your current infrastructure technologies, your staff certifications, and of course, the cost of ownership.

Make sure to understand what features you are looking for and which vendor offers the best solution based on your budget and compatibility with your current infrastructure to avoid painful migration and unexpected issues.

As for the bare metal selection, be certain to choose the proper billing method that fits your budget and your needs. If you require on-the-fly resources that will likely be shut down within hours or weeks, then hourly billing is probably the way to go. However, if you are looking to deploy a steady workload that will run for multiple months, then monthly billing is likely the best choice.

As for the hardware itself, leverage the hourly billing of bare metal to test your application on the server of your choice, and run tests for a few days or weeks until you find the right bare metal configuration for your performance requirements.

Bare Metal or Hypervisor? Making the Choice

In the end, there is no single story for the bare metal server with native workload versus bare metal with a hypervisor and virtualized workloads. Both bare metal and hypervisor have their advantages, and nothing would stop a small development firm from leveraging bare metal servers with hypervisors. It simply is a matter of selecting which technology best fits your needs, and what you feel most comfortable with.

Start with a proof of concept with an hourly bare metal server (with a hypervisor or running your application natively on the bare metal), and see how your application performs. Evaluate the impact on your infrastructure and service management, and move forward according to your findings and experience. There’s nothing like giving it a spin to get the feel of it!

Explore HorizonIQ
Bare Metal


About Author

Paul Painter

Director, Solutions Engineering

Read More
Jul 30, 2013

Bare-metal cloud use cases: Let’s get physical

Ansley Kilgore

The cloud hasn’t always been the ideal choice for performing large, resource-intensive workloads. Scenarios that require high disk I/O and large amounts of compute resources are generally better suited for physical, dedicated servers. While the automation and self-service capabilities of the cloud are valuable, the virtualization layer can take up space that could be used to process your workload. The bare-metal cloud offers a best-of-both-worlds approach – a dedicated environment without the overhead of virtualization, but without sacrificing the flexibility of the cloud (tweet this).

Bare-metal servers do not run a hypervisor and are not virtualized, and can be delivered on a cloud-like service model. This balances the scalability of the cloud with the performance capabilities found in monthly dedicated server hosting plans. The hardware is fully dedicated to the customer, including any additional storage that may be required. Bare-metal instances can be provisioned and decommissioned as needed, providing access to high-performance dedicated servers on demand. Depending on the application and use case, a single bare-metal server can support greater workloads than multiple similarly sized VMs.

Bare-metal servers work well for high-performance use cases such as media encoding and render farms, operations that require short-term data-intensive functions without latency or overhead delays.

Use Case: Media Transcoding
Many sites with user-generated content need media transcoding for publishing to an origin-store Content Delivery Network (CDN). When a user uploads a video, the site needs to transcode the video file into a common format viewable by site visitors. The transcoding software for both audio and video are processor-intensive, and if located on the same machine as the web server or used in a multi-tenant environment, performance may suffer. While a site can send the original file to a remote server for processing and have dedicated resources available to the task, a traditional monthly dedicated server may sit idle during slow traffic periods or be overextended during peak traffic. Bare-metal cloud allows this site to take advantage of the cloud-like flexibility of on-demand provisioning while maintaining the dedicated processing resources of a physical server.

Use Case: Render Farm
Many commercial 3D animation and CAD software support a “render farm” mode, where a regular desktop workstation can be turned into a node in a rendering cluster. This was used by animation companies to develop original media assets during the day, and turn their workstations into a rendering cluster after hours to process their files. With bare-metal cloud, a designer could maintain a single always-on “master” node to submit rendering jobs. This master node would then interact with other hardware nodes for processing the individual frames that need to be rendered. These hardware nodes could be provisioned by design staff as needed to process large or small jobs, or the master node software could be adapted to provision additional instances as needed through a provisioning API.

By combining non-virtualized, physical resources with the service delivery and automation capabilities of the cloud, bare-metal servers can focus all their computing power on your use case. This increases your operational efficiency and provides better performance for smaller investment.

Explore HorizonIQ
Bare Metal


About Author

Ansley Kilgore

Read More