Choosing analytics: built-in, on-premises, or cloud-based

With the announcement of Tintri Analytics, we delivered on our vision: providing comprehensive, application-centric real-time analytics (using fully integrated on-prem and cloud-based solutions) that provide predictive, actionable insights based on historical system data (of up to 3 years).

Customers can now automatically (or manually) group VMs based on applications to analyze application profiles, run what-if scenarios, and model workload growth in terms of performance, capacity and flash working sets.Analytics

When you consider a storage solution refresh, analytics probably tops your list of needed features. It simplifies IT’s job, makes IT more productive and helps organizations save time and money.

The question is, what type of analytics should an organization look to have—built-in, on-premises or cloud-based? If you are just getting started, any sort of analytics would be great! Most storage vendors have an on-premises and/or a cloud-based solution. But an ideal storage product should have all three, as each of them has its own irreplaceable use case. Let’s take a look at each one.

Built-in analytics for auto-tuning

Built-in analytics that the system uses for self-tuning are uncommon in the industry. Tintri’s unique auto-QoS capability is a great example that uses built-in analytics, available at a vDisk level, to logically divide up all the storage resources and allocate the right amount of shares of the right type of resource (flash, CPU, network buffers etc.) to each vDisk. By doing this, a Tintri VMstore ensures that each vDisk is isolated from the other, without noisy neighbors.

Operationally, this simplifies architecture, as the IT team doesn’t have to figure out the number of its LUNs/volumes, the size of its LUNs/volumes, which workloads would work well together and so on. It can focus on just adding VMs to a datastore/storage repository as long as it has capacity and performance headroom available (as shown by the Tintri dashboard).

On-premises real-time analytics

On-prem analytics are extracted from a storage system by a built-in or external application deployed within the environment. Admins can consult these real-time analytics to help troubleshoot a live problem or store them for historical information. Admins can further use these analytics to help their storage solution deliver a prescriptive approach to placing workloads, and provide short-term historical data for trending, reporting and chargeback.

Tintri VMstore takes advantage of its built-in analytics to deliver an on-prem solution for analytics through both the VMstore GUI and Tintri Global Center. Up to a month of history can be imported into software like vRealize Operations, Nagios, Solarwinds and more.

Of course, customers don’t have to wait before they can see these analytics—unlike with cloud-based analytics, they can monitor systems in real-time.

Cloud-based predictive analytics

Cloud-based analytics help customers with long-term trending, what-if scenarios, predictive and comparative analytics. But not all cloud-based analytics are created equal. Some just show the metrics, while others let you trend storage capacity and performance. But the majority of them can’t go application-granular across multiple hypervisors, especially in a virtual environment. They’re just statistical guesswork based on LUN/volume data.

And that’s where Tintri Analytics separate themselves from the pack. With a VM-Aware approach, we understand applications, group them automatically and provide great insights across customers data.

Your IT team wants to be proactive, working on solving business problems instead of doing day-to-day mundane tasks. That’s why each of these three categories of analytics are must-haves. With Tintri Analytics, Tintri’s committed to reducing the pressure on storage and system admins, and helping to grow, not stall, your organization.



What’s New in All-Flash?

Today, Tintri announced the Tintri VMstore T5000 All-Flash series—the world’s first all-flash storage system that lets you work at the VM level—leading a launch that includes Tintri OS 4.0, Tintri Global Center 2.1 and VMstack, Tintri’s partner-led converged stack. Since its inception in 2008, Tintri has delivered differentiated and innovative features and products for next-generation virtualized datacenters. And we’re continuing the trend with the game-changing All-Flash VM-Aware Storage (VAS).

Other all-flash vendors claim all-flash can be a solution for all workloads—a case of “if all you have is a hammer then everything looks like a nail.” Or, they’ll argue that all-flash can augment hybrid deployments, with the ability to pin or move entire LUNs and volumes.


But not all workloads in a LUN or volume may have the same needs for flash, performance and latency. So just as we’ve reinvented storage over the past four years, Tintri’s ready to reinvent all-flash. Here’s how:

  • No LUNs. Continuing the Tintri tradition, the T5000 series eliminates LUNs and volumes, letting you focus on applications. We’re welcoming VMs to the all-flash space across multiple hypervisors.
  • Unified management. Aside from standalone installations, the T5000 series can also augment the T800, and vice-versa. Admins can now manage VMs across hybrid-flash and all-flash platforms in a single pool through Tintri Global Center (TGC), with full integration.
  • Fully automated policy-based infrastructure through TGC, with support from vDisk-granular analytics and VM-granular self-managed service groups.

With access to vDisk-granular historical performance data, SLAs and detailed latency information, customers can decide which workloads can benefit from all-flash vs hybrid-flash—especially when our hybrid-flash delivers 99-100% from flash.

But we hear you, storage admins: you want to go into the weeds. Surprise—we’re happy to help. Here’s what else the T5000 series can offer you:

  • Space savings from inline dedupe, compression, cloning and thin provisioning.
  • NVDIMMs, NTB, 10G and more of the latest hardware advancements.
  • Enterprise reliability exceeding 99.999% uptime.
  • Scale of up to 112,000 VMs, 2.3PB and up to 5.4M IOPs (random 60:40 R:W, 8K) in a single TGC implementation.  (These are real-life numbers, not 100% read numbers.)
  • VM-granular snapshots, cloning and replication.
  • vDisk-level Dynamic QoS to eliminate noisy neighbors and ensure peak performance.
  • VM-level Manual QoS to setup performance SLAs through Min and Max IOPs.
  • vDisk (VMDKs, VHDs)-level data synchronization across VMs for test and dev or any operations requiring periodic copying of data.
  • VM-level replication, backup and transfer between Hybrid Flash and All-Flash systems.
  • VM-granular performance analytics with end-to-end latency visualization that includes host, network, storage, contention and throttle latency.

Today, Tintri continues our solid roadmap of business-relevant innovations in storage for modern workloads. We changed the game for hybrid-flash—and we’re doing it again for all-flash.


Why all the modern Storage QoS Implementations are not good enough?

Storage QoS is starting to become a key functionality today. It is mainly driven by a drive towards increased resource utilization and therefore resource sharing. In the past, Storage would have dedicated RAID group for LUNs and the applications would have LUNs dedicated to it. This worked really well in terms of guaranteed IOPs and Isolation. As disk sizes increased and storage technology matured, we started sharing the drives amongst the LUNs and these LUNs were no longer Isolated from each other. Virtualization made the situation even worse because not only the disks got shared amongst LUNs, the LUNs themselves got shared by multiple workloads (VMs). This resulted in Noisy Neighbor problems that impacted these LUNs and Volumes based storage systems.

A few Storage Vendors have some form of manual QoS functionality built-in the storage OS. Tintri for example, from the beginning, built an architecture that enabled an Always ON, fully automatic, dynamic QoS at an individual vDisk level (think VMDKs, VHDs etc.) that ensured automatic storage resource reservation at the vDisk level (based on our built-in IO analytics engine) so that every vDisk gets the performance it needs at sub-ms response time. The architecture is designed such that a new vDisk gets its performance from free reserves available in the system. So, at no point another vDisk that needs more performance impacts an existing vDisk. The approach is different from traditional approach where one manually sets up QoS at a LUN/Volume level but is highly effective for IT organizations that don’t want to hand hold the storage system. Tintri is the only storage product out there that has an Always ON, dynamic QoS enabled within all its storage appliances.

Screen Shot 2015-04-06 at 4.42.02 PM

Having said that, setting up QoS manually does have its play in Service Provider (SP) Space as well as some Private Cloud implementations where the SP (Public or Private) doesn’t want to give everyone unlimited performance. These SPs want to be able to sell let’s say Platinum Service to their customers and do it dynamically on the fly without even moving the workload. So, coming back to my original point about QoS implemented by Storage Vendors today. Here are the reasons why I say they are not good enough –

Granularity Challenge

In today’s datacenters, workloads are Virtual and Clouds are not implemented without Virtualization. In these Virtualization enabled datacenters, dealing with LUNs/Volumes is a pain. LUNs were brought into the industry 30-40 years back, when the workloads were physical and we started using these LUNs/Volumes even with Virtualized workloads because thats what the storage systems knew. In a virtual environment, a LUN has multiple workloads running in it. Implementing QoS on LUNs has no advantage for virtual workloads whether it is being implemented for isolation or chargeback.  VVols would change it (only for vSphere) but there is still a long way to go there as VVols don’t support all the vSphere features and not all vendors have a practically deployable implementation.

The result is that VMs in a LUN/Volume end up sharing the IOPs limits set up at the LUN/Volume level and therefore end up interfering with each other.

The IOPs Dilemma

Storage QoS is implemented by IOPs. One can combine it with MB/s but only a few vendors allow you to do that. Usually, it is just one or the other.

Now here is the problem, IOPs can have different meaning based on the block size. If I am limiting a LUN/Volume to a 1000 IOPs here is what it could mean –

4K Block size means 4MB/s

8K Block size means 8MB/s

64K Block size means 64MB/s

The same 1000 IOPs can mean 16x more load on a system when looking at 64K block size Vs 4K. That is a lot of difference for a Service Provider to take into account when deciding the pricing for a service. In some cases even large number of small block size IOs may impact storage more than large block IOs. Now some vendors can combine the IOPs limit with throughput to get around this to some extent but ideally service providers want one unit to bill against and want a single scale to measure everyone. Microsoft’s implementation of Normalized IOPs is a great example of such a metric.

The Throttling Effect

Some Storage systems using QoS on LUNs have this problem to deal with specifically when it comes to hosts that have more than one LUN coming from a storage system mapped through a HBA. When one implements QoS limits on a specific LUN and that LUN tries to go above that limit, it gets throttled by the storage system. The IOs get queued up at the HBA level and at that point the host starts to throttle the IO to the LUN and it does that for not just the LUN in question but to all the LUNs coming from that storage, thinking that storage system is not able to take the load that it is trying to send. This makes it practically impossible to implement QoS at an individual LUN level without impacting other LUNs.


The Visibility and Analytics Challenge

Most of the storage vendors have QoS more as a check box with a very few real world deployments. The reason is that QoS is really complex to implement and there are more chances of getting it wrong than right. QoS has to be implemented like a strategy and across all workloads. The challenge is that once someone gets it wrong it is not easy to fix and requires involving Vendor Support teams to determine the cause. Some vendors sell Professional Services around this, which makes it a really expensive feature to implement.

The other point being that QoS itself can become the cause of latency either because of the Max Limits setup on a workload or because of the contention resulting from the cumulative Minimum guaranteed IOPs set up on various workloads exceeding the overall performance capability of the storage system. Ideally the storage systems should give more insight into QoS and its impact on various workloads so that if someone complains of latency or drop in performance, the IT team is quickly able to pin point the reason. None of the storage vendors provide advanced user friendly analytics for QoS today and that is one of the biggest inhibitors in terms of real world adoption of QoS.

To summarize, QoS offered by storage vendors today is not granular enough, it doesn’t have a single scale to measure or apply QoS guarantees/ limits, it doesn’t ensure performance fair-share, it doesn’t guarantee isolation and storage providers don’t have the necessary analytics associated with it to make it easy to implement and then troubleshoot QoS related issues. I think its time to address these challenges so that QoS can be widely accepted and implemented in the datacenter.



FY16 , VVols and Tintri’s Financial Differentiation

vSphere 6.0 is GA and Tintri announced support for vSphere 6.0, VVols and VMware Integrated Openstack along with a plugin for vRealize Operations (vROPs) at PEX this week. We also finished our FY in January and will be off to our Sales Kick Off next week. FY15 was a great year with tremendous growth and record QTRs for Tintri. Tintri continues to lead the way with a product designed from the grounds up for both Flash and Virtualization. With vSphere 6.0 we would bring all the goodness that customer’s love about VMstore to VVols including some of the key differentiators that would separate us from the pack –

  • 99-100% IO from flash
  • VM granular operations
  • VM granular visibility and latency visualization
  • Per VM/VVol analytics
  • Automatics Per VM partitioning of Storage resources (Performance Reserves) based on the analytics
  • QoS and Performance Fair share at a VM level (there are a lot more exciting things coming in this space. Stay Tuned for an update on this….)
  • Latency Visualization across the infrastructure (Host, Network, Storage). We are going to add more to this in the coming weeks … stay tuned
  • 1M VVols per VMstore
    • With VVols implementation, a VM may need as low as 5 VVols and as high as 100s of VVols (with snapshot, clones etc.) and a 1000 VM install would require 10s of Thousands to 100s of thousands VVols.
  • VM Granular Automation
  • Ability to Manage, Monitor and analyze up to 112,000 VMs from a single pane of glass

The VVol race that the storage Vendors are starting now was won by Tintri four years back. If you would be evaluating VVols in the coming weeks/months, you should definitely read this blog here to understand what you should ask from your storage vendor when it comes to VVols.

Tintri from the beginning focused on using software to drive innovation and one of the key differentiators about the technology is its ability to deliver 99-100% IO from flash, which is driven by our software unlike all-flash vendors that use Brute Force to deliver performance. The advantage is that Tintri can address a broad spectrum of workloads at a very low cost.

Workload Breakup

What it means is that unlike an All-Flash solution where $/IOP is low and $/GB is high or a hybrid solution where $/IOP is high and $/GB is low, Tintri brings both $/GB and $/IOP at low levels without over depending on Space Savings (Dedup & Compression), therefore delivering a better $/workload at a very high density.


Our focus on Virtualization continues to help us differentiate and bring new Virtualization & Cloud Centric functionality faster to the market. The result is a platform that is 5-10x cheaper on CAPEX, 60x cheaper on OPEX and highly scalable.



Simplifying Storage Chargeback/Showback with Tintri – Part 2 (Capacity)

In the first post of this three part series, we discussed the challenges around Storage Chargeback and Showback in traditional environments. This post will focus on the Capacity based Model for Chargeback/Showback and how Tintri brings in more accuracy and value add to the model.

As we all know (by now – refer my other blog posts), Tintri doesn’t use a LUN/Volume abstraction layer like traditional storage platforms. LUNs/Volumes were designed for physical environments and we just continued to use them for virtualized environments since there was no innovation done by storage vendors specifically for virtualized deployments (until now). Tintri uses vDisks (think VMDKs, VVOLS, VHDs etc.) as the abstraction layer in the storage platform. Using vDisks as the abstraction layer, allows it to see at the right level of abstraction without the added complexity or layers of LUNs/Volumes. What this means to a Service Provider (Internal IT or Public) is that now they can not just look at what is provisioned to a VM or used inside the VM but the overall Capacity Footprint of a VM. The overall Capacity Footprint of a VM not just consists of the Live Data but also space used for other things like Data Protection.

In the example below, I have highlighted the VM Demo-A. As you can see that in the Tintri GUI, you not only see the provisioned space but also the actual used space by the VM in the Used GiB column. Double clicking the VM shows us various graphs for the VM and in this case I have selected the ‘Space’ Graph that shows us exactly how a 50GiB VM is using 134.5GiBs. We breakdown the 134.5 GiBs into Live Data, Hypervisor Snapshots and Tintri Snapshots to give a complete picture of the Capacity Footprint of the VM.

Capacity Showback

Tintri GUI Screenshot showing the Capacity Footprint of a VM

This is not only important because it allows the IT/Service Provider to charge for the right amount of storage consumed by the tenant, therefore increasing the accuracy and predictability around Storage consumption but also because now the IT/Service Provider can provide more insight and value add.

Chargeback/Showback models can be complex. Here we take a very simplistic example –

  • If we consider the cost of 1 GB of storage as $3
    • With a traditional storage the chargeback would be 50GB x $3 = $150
    • Whereas the actual cost of the VM Demo-A is 134.5 GB x $3 = $403.5
      • So either the Service Provider is taking a hit here or is including this cost as a buffer in the overall costing per GB, making it less competitive
    • As a value add, IT/Service Provider can show the various buckets in which the capacity is being utilized in order for the Tenant to reduce its cost
      • So in case of the VM Demo-A the breakup would be –
        • Cost of Storing Live Data – 45GB x $3 = $135
        • Cost of Data Protection – $3 x (9.29GB+80GB) = $267.87
      • With this information in hand, the tenant can take a decision on deleting some of the older snapshot copies to reduce the cost of running the VM from a capacity standpoint

As we can see in this example, Tintri brings in more simplicity and accuracy with a Capacity based chargeback/showback model . In the next post, I will discuss the performance based model and how Tintri can help with implementing something that is potentially more accurate than a typical IOPs based model.

Thanks for reading.



Simplifying Storage Chargeback/Showback with Tintri- Part 1

Enterprises today are moving towards a Private Cloud based model for running their IT Infrastructure or they are looking at Public Cloud for some of their select workloads. One of the key requirements of deploying a Private Cloud or consuming the public cloud is Chargeback or Showback.chargeback

As we know, with Chargeback, the IT Department/Service Provider hands over a formal bill to the Line of Businesses (LOB)/Customer to recover costs of delivering the service to them whereas with Showback, there is no exchange of any Bills or money.With Showback, IT just tries to introduce a culture of Cost Awareness, Cost Justification, Capacity Planning and Awareness. The model chosen by different enterprises depends on policies and objectives that different Enterprises have.

Chargeback/Showback – The Storage Challenge

Both Chargeback and Showback require some definitive metrics at various levels based on which the overall cost can be measured for delivering a service. Here the discussion will focus on the Data Storage side of the equation.

Continue reading

Why Tintri’s milestone of 200,000 VMs and 16PB of Storage is bigger than you think?

Yesterday we announced a significant milestone of 200,000 VMs deployed on Tintri with close to 16PB addressed by these VMs.

Now, these numbers may not look big on their own, but they become really interesting when you combine it with the fact we have doubled the number of VMs in last 6 months and that capacity numbers are for VMs experiencing sub-ms performance. All of this has come at the cost of incumbent storage providers where Tintri beat Modern & Traditional Storage Vendors in process.

Tintri brought a revolutionary Storage platform in 2011 designed for Flash just to run VMs and vDisks. The idea was to have a storage platform that was designed with Virtualization in mind and not Physical Workloads. To date, Tintri VMstore is the only external, best-of-breed platform that has laser focus on Virtualization. The General Purpose Platforms (Traditional & Modern, Flash & Hybrid) continue to focus on Physical workloads even as the customers are moving towards 100% virtualization.=

Where Tintri continues to have edge across multiple Hypervisors is the vDisk level visibility of workloads as compared to General Purpose Platforms that have no idea about what happens at the Hypervisor layer.

One of the drawbacks of this is the IO blender effect that everyone talks about with no effort to provide any solution except throwing flash at it. Tintri is the only platform that addresses it, delivering high performance, sub-ms response times, vDisk granular isolation, predictability, analytics across the infrastructure and per VM data services.

VMware VVOLs is the step in the right direction but it is just an enabler for one hypervisor and it’ll depend upon a vendor on what functionalities it brings to the storage platforms. With all our experience in managing and understanding VMs, Tintri is committed to bring all the goodness that the customers have experienced on the VMstore and much more to VVOLs. And only Tintri would be able to provide customers a choice of deploying VMs either with VVOLs or with an open implementation like NFS without compromising on functionality.

How will we maintain this growth, as the base gets bigger?

Tintri has a solid roadmap. Our focus on Virtualization allows us to bring better functionality around Virtualization & Cloud, much faster to the market. We don’t have to try to be good for all workloads as other General Purpose Platforms and in process compromise everywhere. We are the best Storage Platform for Virtualization & Cloud and would continue to be so for a long time.

If you want to experience all-flash performance for all your virtualized applications (not just the critical ones) at 1/4th of the cost of an over-provisioned flash array (through over-sold dedup and compression) with all the analytics, auto-tuning, VM granular goodness and isolation that our customers are experiencing, I would encourage you to contact your Tintri Rep or your preferred partner. If you plan to attend VMworld, we are going to have big presence there. You can book a Private meeting with Tintri or attend one of our sessions. See details here.