byaluciani01-24-201711:07 AM - edited 01-24-201711:11 AM
This post authored by Dan Chilton, Sr Solutions Performance Engineer at Nutanix
Nutanix recently released version 5.0 of the Acropolis Operating System and with that release came an exciting Nutanix product feature—Acropolis File Services, or AFS, for short. This new feature deploys like an app on the Nutanix platform. It provides the ability to deploy a robust, enterprise-capable SMB file sharing solution on a Nutanix cluster. By providing this feature within the Nutanix platform, it removes the need to maintain a separate stand-alone network attached storage (NAS) solution for Windows / SMB file services.
AFS is a clustered, distributed file server that runs as a set of file server VMs on the Nutanix platform. AFS has performance scalability built into the flexible design. As storage and performance needs grow, AFS can easily scale out or up, while also allowing dynamic load balancing. For more of a deep dive on AFS features and architecture, read ourtech note.
The following workloads are well suited to be supported by the Nutanix designed AFS.
Windows user home directories
Virtual desktop user remote profiles
In this blog, we discuss file server performance requirements and testing methods, as well as how AFS meets and exceeds these requirements.
One of the first things that customers want to know when considering a file server solution is how well it performs. As someone who focuses on solution performance in my day job, I know how important this is. However, to answer the question, we first need to understand what the customer wants to do with the solution. The customer interactions with the file server constitute the workload.
Here are some example workloads:
Copying one large file at a time to the file server.
Browsing a list of files in a directory and then reading the desired file.
Downloading a file from the file server to the local client desktop.
Editing a document with Microsoft Word and then saving the document.
Multiple users logging on to virtual desktops that have their remote profiles stored on a file server
We used tools throughout the development and release cycle to test the performance of these workloads and others to make sure that we would perform well for our target use cases. To test workload #1, we used robocopy/copy and paste, #2-#4Microsoft File Service Capacity Tool or FSCT, and #5LoginVSI.
Single File Copy Test = Poor Test of Performance
File copy is easy to test (just copy and paste a large file) but it is a poor indicator of file system performance due to a low number of outstanding I/O and the single threaded nature of the workload. This means that a file copy test doesn’t accurately test what the fileserver can do.
Typical real world situations have many clients or applications requesting data concurrently driving a workload. Don’t just take my word for it, check out this discussion from Microsoft OneDrive team member Jose Barreto.
FSCT is a performance test suite based on the analysis of real user home directory operations at Microsoft. It has been used to prove performance by file server vendors including Microsoft, NetApp, and EMC. Some key aspects of FSCT that make it so valuable for simulating customer user home directory environments are:
It tests Active Directory integration, including Windows Domain Controllers, clients, user accounts, authentication, and permissions
Users connect to their home directory with multiple sub-directories, for a total of 270 files/folders and ~80MB of data
Users execute scenarios that create file service metadata workloads
Operations include cmdline file download/upload, Windows Explorer file delete, drag/drop, MS Word file open/close, and save
Throughput is measured in user count of concurrent sustained users, instead of IOPS/FileOPs.
We found that AFS provides a solution for home directories that can be scaled up as user count grows.
With FSCT, we pushed our file server to the limits and established reliable user connection counts per file server VM (FSVM) node that can be scaled up as needed.
We translated this data into the Nutanix Sizing tool to provide quality sizing proposals coupled with room for future growth.
The chart below shows the capability of a single AFS file server node in terms of concurrent heavy user workload.
The AFS solution starts with as few as three VMs, but we have successfully tested a solution scaled out to as many as 16 AFS virtual machines on a 16-node cluster. AFS functionality can be added to existing Nutanix clusters to leverage extra storage capacity or deployed as a standalone file server cluster.
You can scale out small four vCPU and 16 GB RAM AFS VM nodes to support thousands of users by distributing the load across AFS nodes.
We configured LoginVSI, a popular, industry-standard VDI sizing tool, to store the virtual desktop user remote profiles on the AFS solution. We also used Citrix Profile management, a robust enterprise VDI deployment suite. The workload includes Windows OS operations and applications including Microsoft Outlook.
With VDI profiles, the AFS server consumes cycles mainly during the boot phase; once the desktops are booted, they consume most of the CPU and RAM cycles. Accordingly, to gauge AFS performance, we focused on average logon time as the key metric (of course, less is more when it comes to logon time). We wanted to see if logon time increased with clients connecting to AFS file shares to read their profiles at boot up, rather than to the local c:\drive.
We found that we could easily support a 400 virtual desktop deployment with our smallest AFS cluster. As shown in the chart below, the total user desktop logon time was 7.2 seconds for local c:\drive and 6.8 for remotely stored profile. (Note that these logon times are typical for LoginVSI virtual desktops and should not be confused with individual I/O response times, which are often in the ms and us range.)
Why Choose AFS? Performance Scalability by Design
Scale Out—Clustered by design with at least three nodes, AFS currently supports up to 16 nodes. Workloads can be distributed evenly across small or large Nutanix clusters. You can add more file server VMs as additional storage or compute are needed.
Scale Up—Recognizing that some file services workloads require large amounts of processing (CPU) and caching (memory), AFS nodes can scale up as requirements grow by adding additional vCPUs and RAM.
Load Balancing–As file data grows, the storage or processing demands sometimes cause a hot node. AFS easily solves this by rebalancing data and processing across nodes.
Analytics-driven—A fine-grained analytics engine built into AFS continually analyzes the storage and performance consumption of the file server. From this analysis, AFS recommends scaling out, scaling up, or rebalancing. This feature can reduce TCO by minimizing administration time and performance troubleshooting.
Our testing shows that the flexible, clustered design of AFS can provide the performance and scalability enterprises need for the most demanding SMB file sharing environments, all without the added expense of a standalone NAS appliance. Talk to your Nutanix partner to request a demo for AFS and leave the file storage and management to us.
If you are new to Nutanix, we invite you to start the conversation on how the Nutanix Enterprise Cloud Platform can work for your IT environment. Send us a note at firstname.lastname@example.org or follow us on Twitter and join the conversation in our community forums.
Disclaimer: This blog may contain 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.
As the XXXI Olympic games came to a close, we were again reminded of the dedication required to compete at extraordinary levels. Athletes such as Usain Bolt, Michael Phelps, and Simone Biles clearly separated themselves from the rest of the pack to be the fastest and the most consistent. They have what it takes to succeed and have won gold medals consistently.
For example, Usain Bolt has won nine gold medals at the last three Olympics, all while only spending a couple of minutes on the track. To achieve this goal, he trained for over 15 years. He did so by pushing himself to get to the gym every day and has a regime that is second to none. His motivation is derived from not wanting to be in second place, enabling him to reach a top speed of 27 mph. We have a similar mantra at Nutanix: we don’t want to be in second place!
We do this through the collective efforts of our broad set of employees with various backgrounds that come to win every day. Nutanix uses consistent innovation and adoption of game-changing technologies such as NVMe, Persistent Memory, and container, file and block services, to push past the norm.
Data continues to expand at exponential rates, and these technologies become paramount in making sense of the massive amounts of data being collected. To fully realize the benefits of these technologies, compute resources must sit near storage resources. This can massively reduce CPU wait times while delivering consistently low latency access to data.
As these technologies mature and become mainstream, choosing the right platform becomes paramount. The Nutanix enterprise cloud platform is built to adapt and take advantage of newer technologies, putting your business ahead of the competition
Sprint the Globe
Sprinting versus running, what’s the difference? A runner maintains an average speed for a longer period, typically not having a great top speed or ever reaching it during a race. However, a sprinter, like Usain Bolt, will accelerate to a top speed in a matter of seconds and will maintain that speed for the entire race. Now imagine if you could utilize the sprinting methodology for the way an application accesses data, but at the duration of a runner. That would be revolutionary!
To understand what these technologies – NVMe and Persistent Memory - can produce, we have to look at where it started. Rotational media, such as Hard Disk Drives (HDD) have been around for a long time. HDDs provided adequate performance for single workloads at very low costs, but as additional workloads are added HDD performance suffers.
HDDs are significantly constrained in performance especially with mixed workloads on the same infrastructure, and for applications like OLTP and OLAP. HDDs have remained stagnant in terms of performance capabilities, and in fact, the I/O density (IOPS/GB) has continued to precipitously decline with each step-up in HDD capacity (e.g. HDD capacities are 6x to 10x higher than they were a few years back, but IOPS have remained at about 75).
HDDs can be viewed as the long distance runner that doesn’t have a high top speed (IOPS). At the same time, flash costs continue to fall as densities increase resulting in Solid State Drives (SSDs) moving into many enterprise datacenters, replacing the long distance runners - HDDs. However, SDDs were built using the same interfaces as HDDs, and can’t reap the true performance flash offers.
As newer solid state and persistent memory technologies are developed, new interfaces must be designed to take advantage of these performance characteristics.
NVMe is the new access standard for PCI express SSDs. NVMe takes advantage of SSDs by being able to read or write lots of data simultaneously, parallelizing access. As an example, 3D-XPoint technology, with the right interface, has made waves on the performance it will offer. Intel and Micron have summarized the benefits of the technology in a document they wrote (included a small excerpt below).
“In the time it takes an HDD to sprint the length of a basketball court, NAND could finish a marathon, and 3D-XPoint technology could nearly circle the globe.”
Looking at today’s options for high-performance workloads, NAND and DRAM are the two storage/memory media that come to mind. Today’s NAND technologies provide low costs and are non-volatility, and DRAM provides superior performance.
If you take the best of these two technologies and combine them, you get persistent memory technology. It is memory-class speeds which are cheaper than DRAM, faster than NAND, and non-volatile like NAND. We believe as this technology makes inroads into the datacenter, you will start seeing it in SSD type form factors.
The culmination of the two technologies enables applications to have unparalleled access to data without bottlenecks. The only way to truly harness these technologies is to converge compute, flash storage and virtualization on the same platform.
Nutanix keeps the server close to the data. In contrast, legacy three-tier architectures separate applications from storage by networks and storage controllers incurring additional latencies. In these legacy architectures, networks will quickly become saturated as these technologies surface.
Multiple Parallel Lanes
Consider a scenario where you can leverage many Usain Bolt type sprinters in a widened lane, each holding a piece of data. They would be able to get data between two different points at blazing fast speeds. Applying this to applications accessing data increases the amount of I/O between the application and SSD. Applications would just scream with performance, minimizing the amount of time your users wait for access.
Flash performance is steadily increasing as demand for faster systems continues to rise. Today’s SSDs can deliver over 100,000 IOPS and with a few of them aggregated, they can quickly saturate storage controllers. If a single array holds 10-20 SSDs on average, then you are probably not getting the full potential of the SSDs.
The same situation applies when looking at the network where bandwidth becomes a concern. A typical enterprise SSD using AHCI (Advanced Host Controller Interface) interfaces can deliver on average 500MB/s read and 450MB/s write performance, again saturating common network infrastructure. SATA drives with AHCI interfaces typically have a single queue and can handle 32 commands.
AHCI was designed to handle slow read/write operations and not designed for SSDs that can handle large amounts of I/Os. NVMe, on the other hand can handle parallel operations, up to 64,000 queues with 64,000 commands per queue, enabling simultaneous disk I/Os.
Keep Pushing the Norm
The modern hyperconverged architectures not only shorten the path (direct PCIe access) but also enable a significant increase in parallelization, which can allow massive amounts of data to be processed in real time.
As you look for a next-generation platform to push mixed application workloads and drive higher business value, consider a solution that can take advantage of newer NVMe and SSDs options as these technologies become mainstream.
Come See Us at Microsoft Ignite 2016
Nutanix is proud to be a Platinum Sponsor at Microsoft Ignite 2016 at the Georgia World Congress Center in Atlanta, September 26 – September 30, 2016. Visit us at our booth and hear how our customers are deploying large-scale environments with Nutanix all-flash solutions.
If you like what you see on the show floor and want to have a more in-depth discussion about our solutions, we would love to talk to you. Visit the Nutanix booth to schedule a meeting with our engineering, product management, or product marketing team.
Let’s Get Social!
You can stay connected with Nutanix throughout the entire Ignite 2016 event by following @Nutanix on Twitter, and connecting with us on the Next community forms (next.nutanix.com)
We look forward to meeting you at our booth #1310 at Ignite in Atlanta where we will show you how to deliver mission-critical business applications securely and reliably, all while reducing your TCO at scale. Your IT organization will truly be elevated to focus on delivering core business value and innovation by making your infrastructure invisible with the Nutanix enterprise cloud platform.
This post is authored by Rohit Goyal, Product Marketing Manager 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.
byaluciani08-17-201609:20 AM - edited 08-17-201602:09 PM
Nutanix has engineered an all-flash enterprise cloud platform, converging compute, flash storage, and virtualization into a forward-looking architecture and simplistic consumer-grade management interface, allowing enterprises to focus on driving business value.
Flash continues to evolve with NVMe and 3D XPoint technologies, significantly improving application response times, storage density, and durability. Legacy platforms built on outdated architectures are not ideally suited to take advantage of these technology enhancements.
In this post, let’s take a look at how Nutanix all-flash solutions are enabling organizations to keep up with exponential data growth, always-on availability requirements, and the lightning-fast performance that today’s applications demand—all without creating infrastructure silos.
Flash and Servers Just Work Together
Over the last several years, flash prices have been on a decline, making it a more cost-effective option for enterprise storage. However, when replacing HDDs with SSDs (Solid State Drives), and creating an all-flash configuration with legacy three-tier infrastructure, there are several limitations that should be considered, including network latencies, bottlenecks, and management complexities.
This inherently limits the application from utilizing the full potential provided by flash. Another concern is network and storage controller bottlenecks, both of which can limit the performance flash can provide. All-flash storage systems typically have dual controllers, and even though they can be active at the same time, the total number of IOPS is limited.
Lastly, management of this entire three-tier infrastructure is complex, especially when working with multiple vendors for all-flash arrays, storage networks, and servers.
With hyperconverged infrastructure, flash is closer to where the virtual machines (VM) live and application latency can be significantly reduced. The Nutanix enterprise cloud platform can deliver the benefits of local flash performance without sacrificing enterprise data services (including dedupe, compression, cloning, and snapshots) through simple scalable building blocks.
As a result, infrastructure complexity is reduced and management can be significantly simplified.
But it doesn’t stop there. The data locality feature that is unique to Nutanix is designed to keep network traffic minimized even as the solution scales, giving applications consistent, low latency, and predictable performance. VMs directly interact with the data without introducing network bottlenecks and without requiring any special tuning by the IT admin.
Choice with All-Flash Platforms and Powerful Software
Nutanix recently announced the availability of all-flash across the entire portfolio of Nutanix hardware platforms*. With a very modest price increase over similar capacity hybrid storage solutions, moving to a hyperconverged all-flash platform is now a simple decision.
Organizations also have a choice when considering Nutanix all-flash solutions, with Nutanix NX, Dell XC, or Lenovo HX platforms. The platforms are configured-to-order, providing choice in memory, compute, and all-flash storage capacity to meet tier 0/1 application requirements. The all-flash configurations enable you to choose up to 48TB of RAW all-flash storage capacity.
Additionally, the Nutanix platform can lead to OpEx and CapEx savings. Power, cooling, and rack space are significantly reduced as separate silo’d three-tier infrastructure is no longer required. Admins can also spend far less time managing and troubleshooting the environment with an easy to use interface.
Prism, the Nutanix consumer-grade management interface, gives administrators a simple and elegant way to manage their virtual environments. Powered by advanced data analytics and heuristics, it simplifies and streamlines common workflows with built-in automation and one-click deployments. Businesses rely on data more than ever, and with Nutanix all-flash solutions, you have the potential to drive increased business revenue and employee productivity.
Customers such as Valpak are benefitting from improved performance of the Nutanix all-flash solution for their high IOPS data warehousing applications and are planning to migrate their Java application servers, WordPress, SQL Server, SAP, and Sybase databases over as well.
Come See Us at VMworld 2016
Nutanix is proud to be a Gold Sponsor at VMworld 2016 at the Mandalay Bay Hotel and Convention Center in Las Vegas, August 28 – September 1, 2016. Visit us at our booth and hear how our customers are deploying large-scale environments with Nutanix all-flash solutions.
Swing by Booth # 2311 and Schedule a 1:1
If you like what you see on the show floor and want to have a more in-depth discussion about our solutions, we would love to talk to you. Visit the Nutanix booth to schedule a meeting with our engineering, product management, product marketing, or executive team.
Now for the Fun Stuff…
We have some very exciting gifts lined up to thank you for coming and talking to us. While supplies last, every theater presentation attendee will receive our very popular water bottle to help stay hydrated in Las Vegas. One lucky winner will go home with a prize after every theater presentation (every 30 minutes). Raffle drawings of travel gift cards will be held daily.
But the best of all is the grand prize drawing for a trip to an exciting destination ($5,000 value). The final drawing will be on August 29th, 2016 and you must be present to win. More details will be available at our booth.
Also, get ready to be ThunDRstuck with Nutanix, and AC/DShe at the House of Blues in the Mandalay Bay Casino. Join us for a night of drinks, nibbles, and rock n’ roll on August 29th, 2016, from 8:00-11:00 pm. The party will be rocking.
Let’s Get Social!
You can stay connected with Nutanix throughout the entire VMworld 2016 event by following@nutanixon twitter, and tagging your photos and tweets with #nutanixatvmworld. Also, you can connect with us on the Nutanix Next Community (next.nutanix.com).
We look forward to meeting you at our booth #2311 at VMworld in Las Vegas where we will show you how to deliver mission-critical business applications securely and reliably, all while reducing your TCO at scale. Your IT organization will truly be elevated to focus on delivering core business value and innovation by making your infrastructure invisible with the Nutanix enterprise cloud platform.
This post is authored by Rohit Goyal, Product Marketing Manager at Nutanix