Smart Storage for Virtualization and Private Cloud
Senior Program Manager, Department of Defense Cyber Range
Tintri challenges convention — what we call the ‘storage quo’. Our VMstore™ product takes a completely different approach to address the underlying root causes of storage pain.
Over the last two decades, companies have embraced virtualization — more than 75% of applications are now virtualized. So, to deliver the best performance, the unit of storage management SHOULD be the virtual machine. Alas, storage providers force you to operate in LUNs, volumes, RAID and more.
Not Tintri. Challenging convention means breaking that paradigm. Tintri is the ONLY storage provider that lets you see and act at the VM-level.
“Issues that used to take days to troubleshoot now take minutes. The best part is that we are able to monitor our mission-critical VMs in real-time directly from the VMstore GUI.”
Dustin Mommen, Systems Manager, Liongate Capital Management
Our density and speed are best-in-class, but we know that those alone won’t fundamentally change the way you manage storage. We architected our file system to offer you direct visibility into your VMs. That replaces guesswork with precision and automation, and eliminates the root cause of storage pain.
When there’s a performance problem on a critical VM, set a QoS policy on it that robs performance from somewhere else. When you need to move that VM or when something changes in the storage environment, re-adjust all the QoS policies to try to maintain performance. Iterate until your head hurts.
Tintri provides VM-level performance isolation and optimization out of the box. It eliminates the guesswork involved in traditional QoS by providing granular visualization of performance resources & contention at an individual VM-level. Do you need to set a min or max threshold? No problem. Simply drag and set min/max IOPS and you’re done. What’s the performance impact of your QoS setting? Simply click on the VM to see the latency contributed by host, network, storage and QoS throttle.
“We currently use Tintri T850 as our Private Cloud solution due to its simple design.”
Patrick Wloch, Sr. Cloud Infrastructure Engineer, Peak10
We appeal to change agents. They know that when you challenge convention you make an impact. Here are three examples of what happens when you can see and act in VMs.
When there’s a latency problem, point fingers at the host, network and storage. Troubleshoot at the LUN level.
Open Tintri. Immediately spot the specific VM experiencing a latency issue. See right there — with zero clicks and zero investigation — a simple visual (with numbers) that shows whether the latency is coming from the host, the network or the storage. Fingers used for high fives, not pointing.
“I don’t have to manage storage anymore. My performance bottlenecks are either gone or easy to spot.”
Mike Schultz, Network Team Manager, Reinhart Boerner Van Deuren
Manually tune your storage or re-schedule system activities as load profiles change to try and avoid quality of service issues. Fear the noisy neighbor.
Assign every single VM its own lane to guarantee quality of service. We’ve built a solution that is smart enough to know how to handle bursts, and to balance load through automated tuning.
“Tintri has reduced the time and effort necessary to manage our storage to ZERO. It just works. I look at the VM-level analytics a few times a day just to be impressed by the amazing performance. It's a life changer.”
Luke Gray, Systems Engineer, Yavapai Regional Medical Center
Address performance problems by investing in high-cost flash or intentionally over-provisioning storage.
Apply Tintri’s secret sauce — Active Working Set — to deliver >99% of IO from flash. That is pure performance at the lowest dollars-per-VM on the market. We can win your business on dollars-per-gigabyte too, but we know that as you add more and more VMs, it’s the former metric that matters most.
“Compared to our previous storage, Tintri VMstore can accommodate twice the IOPS at less than a third of the latency in one fourth the footprint.”
Yens Jimenez Steller, Product Development Lab Manager, F5 Networks
Speaking of density and performance, we’d like to introduce you to the…
Tintri VMstore T800 Series
We’re incredibly proud of our hardware—we’re pretty dense for a smart storage company: up to 100 TB, 140K IOPS and 3500 VMs in just 4U; or, if you prefer, 1PB, 1.4M IOPS and 35,000 VMs in a single rack.
Please check out the VMstore T800 Series specifications below:
|Flasha||8.8 TB||5.3 TB||1.7 TB|
|Rawa||78 TB||52 TB||20 TB|
|Usableb||51 TB||34 TB||11.8 TB|
|Effectivec,d||100 TB||66 TB||23 TB|
|Controller||Gen 5 Dual Controller (active-standby) supporting Tintri OS 3.1 and up|
|Management||2x 1GbE (RJ-45)||2x 1GbE (RJ-45)||2x 1GbE (RJ-45)|
|2x 1GbE (RJ-45)|
|Optional Data||2x 10GbE
|Replication||2x 1GbE (RJ-45)||2x 1GbE (RJ-45)|
|Optional Replication||2x 1GbE (SFP)||2x 1GbE (SFP)||2x 1GbE (RJ-45)
2x 1GbE (SFP)
|Software Functionality||Ethernet failover link aggregation, VLAN tagging, IP aliasing, LACP|
|Dimensions (HxWxD)||4 RU, 7” (178 mm) x 19” (483 mm) x 28.5” (724 mm)|
|Weight||108 lbs (49.1 kg)||108 lbs (49.1 kg)||106 lbs (48.2 kg)|
|Power & Environmentals|
@ 50-60 Hz (nom/max)
|780 VA / 1,000 VA||714 VA / 861 VA||630 VA / 788 VA|
|Watts (nom/max)||740 / 950||680 / 820||600 / 750|
|BTU (nom/max)||2,525 / 3,240||2,320 / 2,800||2,000 / 2,560|
|Power Supplies||Dual redundant hot swappable with a choice of NEMA or IEC plug types|
|Operating Temperature||5°C to 40°C (41°F to 104°F)|
|Non-Operating Temperature||-40°C to 70°C (-40°F to 158°F)|
|Operating Humidity||8% to 90% (non-condensing)|
|Non-Operating Humidity||5% to 95% (non-condensing)|
|Safety||CSA/EN/IEC 60950-1, GOST|
|Emissions||FCC Class A, ICES-003 Class A, VCCI Class A|
|Agency||CE, CSA, VCCI, CCC|
|Administration||Tintri Global Center™, Web (https), KVM (console), SMTP and SNMP for alerts|
|Support||Proactive support with automated phone home and case creation|
|Virtualization Software Support|
|Hypervisor||VMware vSphere 4.x - 6.x (NFS)
Red Hat Enterprise Virtualization 3.3+ (NFS)
Microsoft Hyper-V 2012 (SMB3)
|VDI||VMware Horizon View and Citrix XenDesktop|
|Multi-VMstore||Tintri Global Center™|
(a) All capacity values are based on Base 10 (i.e., 1 TB = 1,000,000,000,000 bytes)
(b) After double parity RAID-6, spare, and system overhead
(c) Effective usable capacity is calculated after data reduction, which provides up to 2.5x capacity saving from compression and deduplication from per-VM cloning.
(d) Effective usable capacity assumes a data reduction ratio of 1.95x, but does not include savings from thin provisioning which typically provides an additional 2.3x capacity benefit (based on actual customer results)