@NutanixNext
Visit Nutanix

Nutanix Connect Blog

Welcome to the Nutanix NEXT community. To get started please read our short welcome post. Thanks!

Showing results for 
Search instead for 
Do you mean 

Configuring Iometer Performance Tests on Nutanix the Right Way

by Community Manager ‎09-27-2016 09:04 AM - edited ‎11-01-2016 12:03 PM (3,074 Views)

On the Solutions & Performance team at Nutanix, we’re often working with customers or prospective customers to help them identify what they really need their hyperconverged infrastructure to do and how to evaluate their platform options based on those specific requirements. There are several benchmarking tools available to help us figure out what a block is doing—and what it could do. When configured properly, each tool shows different aspects of a system’s general performance characteristics in certain scenarios.

 

There’s no such thing as a “one size fits all” performance testing tool. In an ideal situation, you’d want to assess your HCI with actual application workloads rather than by simply generating I/O. Datacenters in real production environments need to provide resiliency and fault isolation, data availability, and consistency as well as fast I/O management.

 

Adding tools such as HammerDB and Benchmark Factory (which generate real database transactions and simulate user activity) or LoginVSI (which performs real desktop work with various applications) to your testing kit can help fill out the picture of how a system would behave under everyday conditions. Of course, the more realistic a test is, the more complicated and time-consuming it can be to set up.

 

With that in mind, sometimes you just need to run Iometer, and that’s fine. Iometer is a convenient tool that uses random data to mimic I/O patterns and sizes to simulate workloads putting stress on storage. As it’s a fairly general-purpose tool, you’ll want to have specific criteria in mind before you begin using Iometer, so you’ll know what the results are telling you.

 

If you’re running Iometer, here are some tips for getting the most representative test results and thinking about them in a real-world context.

 

  • Make sure you’re testing the best node type for your workload requirements. Nutanix offers several different node configurations to meet various customer needs.

 

  • Update your Nutanix software. Acropolis gets better with every release, and running an AOS version below 4.6 doesn’t give you realistic results.

 

 

  • Turn data integrity protections on. Nutanix has checksums enabled at all times and compares blocks on read, because even the fastest system in the world is worthless if it loses your data. Are all the systems you’re testing keeping your data safe?

 

  • Run Iometer with unformatted disks. Using raw disks directly tests the storage controller, so you’re not inadvertently also testing the guest OS filesystem and caching. This approach helps to keep results consistent.

 

  • Test a big enough working set. Reading a small dataset from RAM isn’t a valid measure of performance for applications that exceed the size of the system cache. Few real applications generate I/O sizes of under 4 KB; databases tend to see 8 KB I/O and up.

 

  • Use sustained write tests to look for consistency and predictability in performance over time. Customers running database applications and other critical systems always want to avoid wild variations, which can have a major impact on operational costs and SLAs. To test for stability even under high workloads, try using 512 KB.

 

  • Run tests using random I/O. Most I/O profiles are random in nature, especially as you combine multiple workloads and run them side by side. Testing with random I/O lets you see how well the system provides quality of service and deals with noisy neighbors—aspects of performance that contribute to reliability.

 

  • Keep Iometer’s limitations in mind. While Iometer is great for testing a system’s limits and for understanding how performance responds under different controlled conditions, no actual application performs only one size or type of I/O, and no real production environment runs a steady stream of I/O at full blast all the time.

 

In whatever test suite you use, be sure to look for network bottlenecks as you scale up the test environment. By preserving data locality, Nutanix keeps its network available for service application transactions and ready to respond during maintenance and failure scenarios. Network availability is especially important with newer all-flash technologies.

 

Comparing platform options can be a daunting task, and test results from different vendors and tools can vary widely. We hope that these Iometer best practice tips can help empower you to get the most benefit out of your comparison performance testing.

 

 

This post was authored by Kate Guillemette, Technical Writer at Nutanix, Will Strickland, Solutions & Performance Engineer at Nutanix, and Michael Webster, Manager, Solutions & Performance Engineering at Nutanix. 

 

Disclaimer: This blog contains links to external websites that are not part of Nutanix.com.  Nutanix does not control these sites, and disclaims all responsibility for the content or accuracy of any external site. Our decision to link to an external site should not be considered an endorsement of any content on such site.

The ROI of Nutanix Xpress

by Community Manager ‎09-12-2016 01:49 PM - edited ‎09-14-2016 02:07 PM (4,424 Views)

Picture1.png

 

Smaller and midsized businesses’ days of being treated as second-class IT citizens are over. Nutanix Xpress brings them the infrastructure simplicity of public cloud, but with the in-house capabilities, control and lower costs enjoyed by Global enterprises. Built-in virtualization, superior resiliency and availability, and world-class services and support make Xpress an exceptionally compelling solution.

 

The Problem with Traditional Infrastructure

 

“The simplicity, and reliability that Nutanix as the leading enterprise cloud provider has been bringing to our customers over the last four years has always been relevant to SMB. Nutanix Xpress makes all of the benefits of the Nutanix enterprise cloud platform available in packaging options and price points specifically targeted at these SMB customers. Smaller organizations are now able to leverage the Nutanix Xpress line to power their IT requirements, and compete directly with these larger corporations.”

 

Darren Ashley |Managing Director, BEarena Pty Ltd

 

Xpress was finalist at Best of VMworld 2016 in the hotly contested Virtualization and Cloud Infrastructure category. It is targeted to small and medium sized businesses – typically defined as businesses or organizations with 50 – 500 employees or less than $100M in annual revenues.

 

In order to enjoy the many benefits of virtualization such as reduced hardware and power cost, increased resiliency, simplified administration and so on – these organizations previously had to pay large monies for a diverse set of IT equipment including servers, centralized storage and a storage fabric.

 

Purchasing and setting up this environment was expensive. Managing it was particularly burdensome – often requiring specialized administrators and skill sets along with ongoing training. To make things worse, the business had to purchase separate virtualization and management software along with the hardware to run it.

 

In many cases, resource-constrained smaller businesses could only afford systems that lagged well behind those of their large competitors in terms of resiliency, scalability and capabilities – putting them at a competitive disadvantage.

 

Many smaller organizations have been turning to public cloud as a way to mitigate the challenges of traditional virtualization infrastructure. But cloud is no panacea. It still needs management, is expensive, and requires a monthly ongoing (and often increasing) payment.

 

Setting Virtualization Free

Virtualization has long held the promise of both reducing costs and streamlining IT. But the challenges of traditional infrastructure have kept virtualization from reaching its full potential – especially for small businesses. Until now.

 

Nutanix Xpress finally delivers on the full promise of virtualization by letting businesses focus on apps and services rather than infrastructure. Underlying every facet of the Xpress offering is exceptional consumer-grade simplicity. As just one example, a single click upgrades the Nutanix OS, the firmware, bios and hypervisor non-disruptively in the background.

 

Here are some of the ways in which Nutanix Xpress sets virtualization free:

 

Configuration:  The Xpress solution is “configure to order” and starts as low as US$25K price to customer including three nodes and three years support. It’s a fully, integrated, turnkey solution that eliminates complicated IT Integration, set-up and management.

 

Included AHV Virtualization: The nodes run both VMware and Hyper-V, but come built-in with Nutanix’s next-generation AHV virtualization stack. To date, 60% of Xpress deployments have been using AHV. They can quickly virtualize out of the box: No servers, no storage, no switch fabrics, no separate hypervisor and no separate hypervisor management software.

Installation:  Installation can be done in as little as 60 minutes with server, storage and virtualization all in one appliance.

 

Administration: Nutanix Prism provides an extremely intuitive single pane  interface for managing the entire environment along with rich analytics. A single administrator can manage everything, including AHV, with simple and elegant consumer-grade management. 1-click management, 1-click upgrades, 1-click search, and 1-click issue resolution.

 

Support: World-class support is included for the entire solution: compute, storage and AHV. Nutanix has a Net Promoter Score of +90 and has received the Omega Northface Scoreboard for Excellence in Customer Service the last three years in a row.

 

Picture2.pngData Protection: Integrated data protection and storage backup ensure that critical business information and customer data are never lost. And like everything that Nutanix does, data protection is simple to implement and administer.

 

Cloud Connect: Built-in Cloud Connect provides seamless data backup to AWS or Microsoft Azure.

 

Performance: Xpress enables high-performance delivery of all types of applications, including popular Microsoft applications and Windows Services, infrastructure services (DHCP, DNS, file serving) and general workloads including SQL Server and more.

 

Xpress is powered by the same web-scale technology running in more than 3,100+ organizations across nearly all industries, including retail, finance, healthcare, professional services and more.

 

TCO vs. Traditional Infrastructure

 

The TCO of Xpress vs. traditional infrastructure varies greatly depending upon the configuration, brands and models, but it is not uncommon to see a reduction in costs of 60%:

 

Rack Space: Up to 4 hosts or nodes fit in in a single chassis taking up only 2U of rack space

 

Power/Cooling: Dramatically reduced

 

Administration: Slashed. Xpress practically eliminates all storage tasks including LUN carving, switch zoning, volumes creating, RAID group management, etc. 1-click upgrades enable seamless upgrades of not just the Nutanix OS, but also the firmware and even hypervisor.

 

Hypervisor: Optimal hypervisor can be utilized for the use case. No cost when using AHV.

 

Resiliency: No archaic RAID configurations or single-point-of-failure virtualization management servers to worry about. Self-healing.

 

Analytics: State-of-the-art automation and rich system-wide monitoring for data-driven efficiency. No additional software licensing is required.

 

Scalability: Simply add another node (with a single click) to the cluster

 

TCO Example

 

Let’s look at an example of what a 3-year TCO might look like for Xpress vs. a legacy 3-tier environment (servers + centralized storage + storage network).

 

For this example, we’ll make the following assumptions:

 

3-Tier

 

  • Servers $8,500 each including 3 years support – 900 watts each
  • SAN: $33,000 including 3 years support – 800 watts. No additional switch fabric cost
  • VMware vSphere Essentials Plus: $5,619 for 3 hosts + 3 years support
  • Installation: $10,000
  • Administration: $31,250/year (1/4 of a $125K fully burdened admin)

 

Nutanix Xpress

 

  • Turnkey 3-node solution: $25,000
  • Installation: $3,600
  • Administration: $9,063 (29% of 3-tier – per IDC study)
  • AHV (included)

 

As the following chart shows, 3 year costs for the legacy solution come to $176,771 vs. $61,449 for Nutanix Xpress.

 

Picture3.png

 

This example is conservative; it does not take into account the SAN refresh cost, nor the cost of a forklift upgrade if capacity is reached in less than the 3 year TCO period.

 

TCO vs. Other Hyperconverged Solutions

 

As with traditional infrastructure, the TCO varies greatly. But in many cases, the cost for 3 Nutanix Xpress nodes (and the much greater resiliency that enables) tends to be less than 2 nodes of other solutions. Other items to consider:

 

Simplicity:  Nutanix is the simplest HCI platform to manage (for example, 1-click upgrades) which slashes administrative costs.

 

Virtualization: Xpress includes AHV hypervisor as part of each node. AHV not only eliminates all hypervisor and hypervisor management costs, but also 3rd party software costs such as SQL Server for HA management and specialized analytics packages. It also reduces design, implementation, support, on-going administration costs and business disruption costs.

 

Multi-Hypervisor Support:  Included

 

Replication:  Included

 

Analytics:  Included (HCI industry’s best analytics)

 

Backup to Public Cloud: Included

 

Enterprise Cloud: Most importantly of all, Nutanix considers HCI as simply table stakes for implementing an enterprise cloud. Nutanix is replatforming the datacenter to deliver an invisible cloud-like experience for enterprises of all sizes.

 

An enterprise cloud platform enables business agility benefits such as increased sales/cash flow, decreased downtime, more productive employees, better customer responsiveness, and so on. These business benefits typically trump the infrastructure savings..

 

Picture4.png

 

 

TCO vs. Public Cloud

 

The chart below shows an example of a 3-year TCO comparing a 3-node Nutanix Xpress cluster vs. AWS. We assume 30 VMs will be run in each environment.

 

AWS

 

For AWS, we use on-demand pricing for a standard AWS c4.large EC2 instance which includes 3 vCPUs and 3.8 GB of RAM with a total monthly cost of $138.96. AWS instances require separate Elastic Block Storage (EBS) allocations, and for the purposes of the analysis, we assumed that each VM would use 100 GB of EBS-st1 storage with a total monthly cost of $4.50.  We assume a total connectivity cost at $25/mo. and support costs coming in at $49/mo. for the entire environment. Additionally, we assume no up-front cost to set-up and $607/mo. in administration costs.

 

Nutanix

 

For Nutanix, we assume $25K for the 3 nodes and $3,600 in partner fees to set-up. We assume no rack space cost, but power and cooling costs of $62/month. We assume redundant switches are required at a cost of $4,000 each, and cables at $360. Administration is $755/month.

 

Total 3-year projected costs for AWS come to $159,890 compared to $85,191 for Nutanix. On a monthly basis (ignoring the cost of capital), the cost is $$4,411 for AWS and $2,366 for Nutanix. Of course, after three years Nutanix CapEx is paid off, while the OpEx (1,300% higher for AWS) continues.

 

Picture5.png

 

AHV & EUC

 

Small and midsized organizations including local/regional businesses, educational institutions, local government agencies need solutions that are supported throughout the industry. Nutanix has over 100 partner solutions verified as Nutanix Ready with many of them also certified for AHV.  Microsoft, for example, not only supports all of its software on Nutanix, but even on Nutanix AHV despite having its own Hyper-V hypervisor.

 

Along with Microsoft Windows, desktop and server OSes as guest OSes, Citrix XenApp, and XenDesktop is officially supported on  Nutanix AHV. Running it includes benefits such as management simplicity, scalability, security, resiliency, analytics, and support for the entire stack. Studio-integrated MCS provisioning (starting with XD/XA 7.9) works faster and better with AHV native provisioning and with Nutanix shadow clones.

 

While application virtualization and VDI with traditional infrastructure is very expensive and complex, combining Nutanix Xpress with Citrix makes it extremely simple and affordable. Implementing application virtualization and VDI enables small businesses to save still more money.

 

Connect with Nutanix

Are you ready to learn more about Nutanix Xpress and AHV, including how it would fit in your environment? Reach out to the Nutanix sales team to setup a customized briefing over email: info@nutanix.com or DM us on Twitter @Nutanix. You can also do your own test drive of Nutanix AHV using our complimentary Nutanix Test Drive program whttp://www.nutanix.com/testdrive). Lastly, engage with the vibrant Nutanix community and user groups through our Nutanix Next  (http://next.nutanix.com) community site or through our upcoming Nutanix On Tour event series.

 

This post was authored by Steven Kaplan, VP of Client Strategy at Nutanix

 

Thanks to Steve Conner, Patricia Sampson, Sachin Chheda, David Sangster and Nikita Maheshwari for contributions and editing.

 

See Also

Why Nutanix One-Click Upgrade is One Of The Coolest Features. 04/28/2016. Artur Krdywdzinski. vmwaremine.com.

The Single Biggest Benefit of Nutanix Hyper-Converged Infrastructure (It’s Not What You Think). 12/10/2015. Steve Kaplan. Nutanix.com

7 Reasons Why AHV is The Next Generation Hypervisor. 12/01/2015. Steve Kaplan. Nutanix.com.

 

 

Announcements

One of the features you see a lot of on social media sites are @Mentions. With @Mentions you can acknowledge other community members.

Read More: Did you know: You can @Mention other members
Labels