Continuing on from my previous blog post, detailed in this post is my personal Tintri story. It’s not the typical technology trend, deep-dive or new-feature-focused blog post you’d usually read. It’s personal tale of my journey and will give you some insight to what’s been going on at Tintri, and why last week’s release of TXOS 188.8.131.52 is such a special milestone to myself and many other Tintrians.
I’ve had a number of people ask… why did you return to Tintri? Simple: I love the product! It’s not very often you get a 2nd chance, and here was a 2nd chance… DDN came to Tintri’s rescue. And I want to be a part of it! I didn’t want to let this great technology that I’ve poured in so much time and effort into to simply fade away….. not if I could help it. And that’s why I’m back… to continue helping customers, contribute to the advancement our products, convert non-believers into believers, and make Tintri a household name in the modern Virtualization and Cloud era.
Some of you know me, some of you don’t. Some of you only know me as that crazy geek with the Mohawk you see at VMworld every year. A bit of background: I’ve been working for Tintri for nearly 5 years (Jan 2014), and before that I was an early-adopting, Tintri-loving customer. I’ve been trusting Tintri VMstores to run my VMs since March of 2012. I’m coming up to 7 years of hands-on experience with our product, and I’ve yet to find an equivalent offering in the broader IT marketplace that adds as much value to a virtualized environment as Tintri does.
I became a Tintri customer back when there were ~50 people working at Tintri. I was living in Montreal at the time, working for an offshore financial firm. I was a fanatical customer and loved being included in Tintri’s beta program. I had direct access to engineering… it was awesome! I was a beta tester for v1.4 back in March of 2012, the monumental version that first introduced Tintri snapshots, cloning and multiple vCenters. I actively contributed feedback on betas for many other releases and I was always blown away on how quick and efficiently Tintri could incorporate my feedback to improve the product.
One of my favorite beta test stories goes back to the summer and fall of 2012, contributing to the beta of version 2.0, another monumental release that would bring efficient per-VM a-synchronous replication to the world. I had an Exchange VM with several vDisks attached, including one single large vDisk that weighed in at 1.25TB. I look back at that and don’t see it as big and nasty anymore compared to today’s standards, but you gotta understand the environment to appreciate how big and nasty it was. I had to replicate it over the wire from Cayman to Dublin. The wire available to me was a 20Mbps link, already saturated with other replication traffic, that had had 160ms of latency at the best of times. All previous attempts to make a copy weren’t successful for one reason or another… the line would go down, vCenter would get interrupted, etc. I figured I’d try out Tintri’s new replication and was excited about the WAN-dedupe feature. It was running great… VMs were replicating, and then all the VMs queued up behind my single large VM. But hey, you can only stuff so many bits down a crowded pipe, so I would have to deal with it. But a bigger problem occurred… after 7 days, the replication gave up and restarted. It took quite a while to figure out what was happening, but essentially, the history database in the VMstore is only 7 days and no one expected a single job could ever exceed that time period. Leave it to me to be the outliar!
The talented engineers working at Tintri fixed the concurrency issue so that a single VM wouldn’t breach SLAs of other VMs (especially if they are just tiny deltas), and solved the 7-day limitation. I had hit day 8 and it was still replicating like a champ. I looked at the “Bytes Remaining” graph, and based on the decreasing slope over a 7 day period, I anticipated this job might take 6 weeks to complete. I was off… it actually took 7 weeks! But it completed, and the bits on the other end passed all integrity checks! This replication was now rock-solid and tolerated everything from WAN outages, Windows patching & reboots of the VM (something that screwed up attempts using in-guest tools), vCenter outages, didn’t interfere with our ability to take Veeam backups while it was replicating… in short, it was nothing short of awesome! I had a real-world challenge now solved, and my love for this product and company continued to grow.
By the end of 2013, I had decided I needed to make a move to Silicon Valley and be part of this amazing thing called Tintri. I joined in Jan 2014 as employee number 120 (give or take). I worked my first 9 months remotely in Canada due to some visa complications that prevented me from getting onsite to HQ. Finally, my H1B Visa came through and we moved myself and family (supportive wife and 2 young daughters: 2 & 4 at the time) to Mountain View, CA in Oct 2014. I made it to the mothership and got to meet the many talented engineers whom had created this wonderful technology. Life was awesome!
In August of 2015, we had our first “Tintricity” user meetup the day before VMworld would kick off. It was a great event, and included a special guest appearance by Carl, the Storage Guy and that guy from Silicon Valley:
Many move events and product launches followed over the years. I helped contribute alongside incredibly talented, innovative, and dedicated teammates and we grew to approx. 650 people by summer of 2017. Our customer count was growing, and our customers loved our products almost as much as I did…. some will even claim more!
Over the years, I’d written several whitepapers ranging from XenDesktop to SQL Server to vSphere Best Practices. And I’d have to say that while it may have been one of the most challenging, my Cisco UCS Best Practices Guide may be the one I’m most proudest of. I’ve also delivered dozens of talks and webinars. Without a doubt, my “vNUMA & pNUMA: Save your SQL from Certain Doom-A” VMworld talk, co-presented with Shawn Meyers, takes the cake for crowd favorite. It was the first and only VMworld session to date to get a three-peat due to popular demand, which was back in 2016. We’ve kept updating it and delivered it to packed rooms for 4 VMworlds in a row (even more if you include VMworld Europe & US as separate shows!).
In February 2017, we had a our annual Tintri SKO (Sales Kickoff). Looking back, this was the peak of morale. We felt unstoppable, and customers were excited with what we’ve delivered We were cranking out amazing new features in major releases, the most significant and technically challenging to develop at that time was Synchronous Replication. We were creating VM-centric solutions that were firsts in the industry.
At that SKO, among many other sessions, we had a “Tintri Labs” session that showed off some amazing tech we had proof of concept prototypes of back in the lab. I can’t mention most publicly (yet!!!), but there are two I am comfortable to mention here. Storage vMotion Offload…. Happy to let you guys know this is NOW AVAILABLE to you with our TGC 4.0 release, released on Monday! And another is the SQL-aware demo, where we extend VMstore functionality to natively host MS SQL databases, and provide all the awesome features we have for VMs to databases, such as instant cloning, snapshots & replication. [Spoiler Alert: Watch for this later next year!] Standing on the large stage, looking out at my peers, I demoed this to the audience of very talented Tintri Account Reps, Sales Engineers, Executives & some engineering. The reaction was unanimous… jaws dropped, and then a standing ovation lit up the room. Standing up there, seeing the excitement, and then having all the excited side chats later throughout the SKO was another career highlight. As I mentioned earlier, morale was at a peak high!
In June of 2017, Tintri IPO’ed on the NASDAQ, a milestone many of us had been working towards for so long. I was in Vegas working the booth for Cisco Live!, but at the last minute, I took a red-eye and flew to New York to make it in time for the opening bell. Oops… we IPO’ed a day late! That was an incredible experience and it was one I’ll cherish forever.
There were mixed emotions: negativity stemming from the shock of how much lower our shares went for than what we’d been believing for so many years, and a great sense of pride and positivity… we finally IPO’ed! A MASSIVE milestone for any company!
Employees had a 9 month lockup post-IPO to wait before they could sell any of the options and shares we’d been accumulating for years. March 2018 was supposed to be our long-awaited pay day! Even though there was a reverse-split at the time of IPO and our shares were consolidated 6:1, there was still hope that March could make all our effort and hard-work worth it. When the lockup expired in March, we (Tintri employees) were given our first opportunity to sell shares and exercise options (if they weren’t so far underwater). I won’t get into the details, but needless to say, it was a very “financially inconvenient” time for all of us and stocks were at an all-time low.
Little did I know, that IPO, a milestone we’d been working so hard towards for so many years, was the start of something unexpected. That’s when things started to slide. Many of us were upset (myself included!) at the new value of the options and shares we owned. Some left, seeing that the pay day they’d hoped for could never materialize when March 2018 would finally arrive. Others were let go in several rounds of layoffs intended to strengthen our financial position and decrease our cash burn. But there were quite a few of us that hung on and believed it could only go up for there… how naïve.
By the time the liquidity event (March 2018) would finally arrive, a large number of employees had left and many were on their way out the door feeling financially burnt by a pay off that was well-deserved after many years of intense hard work. If I sold at the very first opportunity, I could have gotten somewhere around $3 share. But I didn’t sell… I felt that the market was being unnecessarily punishing towards us and surely we were worth so much more. But the price continued to slide, and unfortunately that narrow 3 week window employees were allowed to sell in closed. The stock continued to plummet and by the time I would be eligible to sell again, it would be for pennies.
It was a very difficult time for me and those around me. The reality of seeing millions turn into pennies was a shock and incredibly hard to believe for so many of us. We felt like we had been misled and betrayed for so many years. We’d been told how awesome we were doing, watched our internal value of our stock options climb, and it wouldn’t be much longer until we reached cashflow-positive. How could our reality flip seemingly overnight? “How” really isn’t important… all that matters is that it did. Ouch. L
The roller coaster ride was intense! There was a glimmer of hope shortly after March. On April 2nd 2018, a new CEO (Tom Barton) stepped up. He was one of the most impressive leaders I’ve ever had a chance to work for. Inspiring and honest… there was hope! But sadly, it was too little too late. Even the best leaders in the industry couldn’t have turned this situation around… not this late in the game.
I felt it was time to make a change and in May, I left Tintri. This was an incredibly difficult decision for me. I felt like a traitor. As much as I was a believer, I felt like I was giving up too soon. Aside from a passion of all the great Tintri goodness that has launched over the years, there was even more goodness in the backroom that that public had yet to see. For amost 2 years I had been demoing an amazing prototype of some kick-ass SQL-aware storage we were cooking up in our R&D labs, and it was almost ready to go GA. Would I really leave before seeing my baby’s launch later in the year? Sadly: Yes. [NOTE: this kick-ass technology still exists in our labs and WILL see the light of day, but later than anticipated due to obvious reasons].
I left to work for a startup focused on open source S3-compatible object storage. That change of scenery was a breath of fresh air and gave me new perspective. It gave me an excellent opportunity to take off my Tintri blinders that I’d been wearing for so long and take a break from the Kool-aid I’d been drinking for so long. It gave me a chance to shake off the feelings of embarrassment and anger of getting financially burnt. It gave me a chance to work with a new set of incredibly talented and passionate engineers. These engineers gave me a chance to dive deep into Cloud and containers and taught me so much.
I realized that even in a world of containers, VMs have not lost their utility, and are going to remain part of the datacenter landscape for the foreseeable future. Sure, some workloads that run in VMs are better suited to be run as containers, but even nodes where containers run may be better served by VMs than bare metal, depending on the environment (but don’t make me get off track!) Long story longer… while I was gone, I missed the tech… I missed it a lot!
Should I move back to Tintri?
The downfall of Tintri was NOT due to having an inferior product… it was a series of poor business decisions exacerbated by growth expectations and all the pressures that come with being a public company. “Headwinds”, as they say. And once things started to slip and the market bet against us, sadly, it became a self-fulfilling prophecy. This was so difficult for me to comprehend… how can this be happening to such an amazing piece of technology. I learned a valuable lesson… having the best technology isn’t enough!
For better or worse, I watched from the sidelines when bankruptcy was announced in early summer of 2018. It broke my heart. I couldn’t imagine the IT landscape without Tintri being a technology choice for customers in a sea of other tech choices hadn’t changed for so long, at least as far as storage is concerned. It would be too great a loss for this great technology to disappear. Yet here I was, staring into the Abyss, watching this great technology slip away. And along with it, my “Plan B”. After leaving, I figured that if things didn’t work out and/or I wasn’t as passionate about another technology, I could always move back to Tintri. Now that door was closing. We’d been hoping another company would buy us, but none did, and bankruptcy was declared.
But wait! There’s hope! A company I wasn’t familiar with stepped in out of nowhere and made a bid to buy Tintri. Former Tintrians whom had been let go were now getting hired by DDN. I stated to do some research on DDN and was impressed with what I read. While they’ve been focused in the niche space of supercomputing, they were kicking ass in their domain. I got excited and felt hopeful! Perhaps I could return after all? After getting so close to a Tintri-less world, I wanted to get back and help. I wanted to be a part of the reboot…. The Tintriboot! And I wasn’t alone!
The decision to return was almost made, but still took a bit more encouragement. After all, I had been burnt pretty bad! The icing on the cake for me to make the move back to Tintri came at VMworld 2018. I sat in an amazing VMworld session presented by my former colleague Tomer Hagay: Best Practices for Snapshot Heavy Environments (VAP3915BUS) I urge you to watch this video… it is the best material I’ve seen to date on the impact of VMware-based snapshots of VMs, and also does a great job at showing many of Tintri’s strengths in the process. It reminded me just how much I miss interacting with my beloved VMstore UI and how much it continually teaches me about the workloads running within. And I missed my former coworkers… we’d been through so much together, and working with them had always been so rewarding.
Further signs indicating that I should follow my heart & return to Tintri would be revealed to me at the show. I’d pop by the “Tintri by DDN” booth at VMworld to catch up with former colleagues and get the latest on the state of Tintri. Customers kept coming by the booth, having investigated inferior alternatives and were hungrier than ever for more Tintri. “Hopefully the DDN deal goes through soon so I can buy more… I don’t want another storage technology because none of them suit our needs as Tintri does” – a statement I heard many times in one form or another from many of our loyal customers. And then on Wednesday, August 29th, news broke that DDN successfully won the bid for Tintri, and the technology was saved! Not only saved, but is now stronger financially than ever! I knew I had to come back to be part of the Tintriboot, so that’s what I did. I rejoined Tintri on September 17th.
Have I been burnt… yes. BUT, ask yourself, can you think of any technology (or anything else for that matter) that you are so passionate about that you’re willing to give it another shot? Do you believe so strongly in a vision that you’re willing to risk it all again? Clearly, I do. I would strongly urge you to try Tintri for yourself, and then tell me I’m crazy for coming back. You won’t be able to…. Instead, you’ll wonder how you missed out on this amazing tech for so long and wish you’d given it a try sooner.
Fast forward to the present…
It’s coming up to 90 days that I’ve been back. Feels great to call myself a Tintrian again! J Has it been easy? No. But has it been worth it? So far… absolutely! It feels amazing to be part of something as magic as this. Working closely with some of the best professional talent I’ve worked with to date, my team mates never cease to amaze me in their ability to step and make stuff happen! We pulled off 2 lab migrations in this time, and an office move. Teams are being rebuilt and I’m loving the direction things are heading as “Tintri by DDN”. Again, this is why last week’s TXOS 184.108.40.206 release and this week’s TGC 4.0 release are so monumental to me, and many others.
We have a second chance to make an impact and change lives. To bring to admins, architects and dev-ops engineers a real, solid solution that will improve their datacenter lives. Too many companies out there still have never tried Tintri. And they have virtual environments that truly are Tintri’s sweetspot for infrastructure enhancement. And one of the most common reasons I hear is that they may have a few non-virtual physical workloads that won’t run on Tintri. So they look for a vendor with more checkboxes to do more. No, we aren’t a swiss-army knife of storage to serve your every need. But have you ever tried to build a house with a swiss-army knife? No thanks… I’ll stick to specialized best-of-breed tools! And that’s what Tintri is…. a specialized, best-of-breed tool designed to do one thing and do it a hell of a lot better than any other storage platform: run VMs. Expect more from your storage…
What are you waiting for? When would now be a good time to give us a try? It’s up & up from here and we want you to join us on this journey. DDN is a 20+ year old storage company that is fiscally responsible and knows how to successfully run a business. It didn’t become the world’s largest private storage company by accident! I’m very pleased with our new position, and we’re stronger than ever. To the many companies that have stuck with us throughout, THANK YOU! We truly appreciate your business and will continue to work tirelessly to help you prove to your stakeholders that you made the right choice by choosing Tintri. Many more great things to come J And to many of my ex-Tintrian colleagues out there… it’s never too late to come back!
Thanks for listening to my story. Now that I’ve had a chance to share bits about the past, I’m keen to leave it behind us and focus on the future! Stay tuned for many more posts to come, but don’t worry… I won’t bore you again with touch-feely emotional stuff…. I’ll get back to what I do best: geek out! Next post will be much more tech-focused in nature and I’ll walk you through the lab migrations. With a very short amount of time and very few resources, the success of these migrations was only made possible because we had Tintri powering our datacenters. Stay tuned for details, including pics & videos.
Thanks for your time!