Some of the greatest “rocket science” in RISC-V is the simplicity, elegance, and flexibility of it. While some of the individual pieces are not “rocket science,” the collective whole is “rocket science.” It enhances speed-to-innovation, and with advances over the last 15 years in chip design and implementation, this makes it possible for our members to easily use RISC-V for everything from embedded to high performance computing (HPC). Those products will advance many areas including disaggregation, machine learning (ML), storage class memory, etc. Ultimately, it will result in better, faster, more capable end-user products like automobiles, network edge servers, or cloud servers.
As a part of our series about cutting edge technological breakthroughs, I had the pleasure of interviewing Mark Himelstein.
Before RISC-V International Mark Himelstein was the president of Heavenstone, Inc. which concentrated on strategic, management, and technology consulting providing hardware and software product architecture, analysis, mentoring and interim management. Previously, Mark was vice president of engineering and CTO of Graphite Systems, Inc. (acquired by EMC), where he focused on developing large analytics appliances using highly integrated Flash memory. He has also held positions including CTO of Quantum Corp., vice president of Solaris, development engineering at Sun Microsystems and other technical management roles at Apple, Infoblox, and MIPS.
Mark has a bachelor’s degree in computer science and math from Wilkes University and a master’s degree in computer science from University of California, Davis/Livermore. In addition to publishing numerous technical papers and holding many patents, he is the author of the book “100 Questions to Ask Your Software Organization.”
Thank you so much for doing this with us! Can you tell us a story about what brought you to this specific career path?
I have always had a strong interest in architecture and my career went back and forth between executive management and development. The biggest job I had was running Solaris and my biggest task was kick-starting innovation with the team for things like the zettabyte file system (ZFS), DTrace, and Zones. I was able to combine my management experience with technology and have continued to do so since. In December 2019, some of my colleagues from MIPS asked me to consider this job and I said sure. The more I learned, the more excited I was. RISC-V is the heir apparent from earlier RISC processors and the community needed some help in the leadership department, so it was the perfect combination. Over the last seven months since I accepted the position, my interest has only grown. I am grateful to be a part of this effort.
Can you share the most interesting story that happened to you since you began your career?
I have been at this for about 40 years and I am not sure I can pick one! A recent one was from a company I started in 2012 called Graphite Systems that was acquired by EMC. We made a highly parallel flash-based compute appliance for big data. We were working with a social media company doing a database query that required 200 machines in order to run and it typically took 30 minutes to complete. They needed the query done faster. With a 4-socket Intel based OTS server and our parallel flash device attached, we could do the same query in eight minutes! When asked how we did it, the answer was quite unique and elegant. We threw out complexity and used plain text columnar files with basic statistics, on top of highly parallel flash storage (think non-volatile memory host controller interface specification “NVME” on steroids). Everyone had kept building layers of software, which included code people had forgotten were even in the stack. The same layers that enabled rapid development using common building blocks had just become an albatross. Sometimes you have to start from the fundamentals and be simple and elegant. That is what RISC-V is as well; simple and elegant. Designed for flexibility to last for 50 years or more.
Can you tell us about the cutting edge technological breakthroughs that you are working on? How do you think that will help people?
As discussed above, some of the greatest “rocket science” in RISC-V is the simplicity, elegance, and flexibility of it. While some of the individual pieces are not “rocket science,” the collective whole is “rocket science.” It enhances speed-to-innovation, and with advances over the last 15 years in chip design and implementation, this makes it possible for our members to easily use RISC-V for everything from embedded to high performance computing (HPC). Those products will advance many areas including disaggregation, machine learning (ML), storage class memory, etc. Ultimately, it will result in better, faster, more capable end-user products like automobiles, network edge servers, or cloud servers.
That being said, we get the benefit of history. So if you take a look at extensions like vector or crypto, you will see they are “rocket science” and they enable things like ML or security in fundamentally better ways.
How do you think this might change the world?
When I was at MIPS, we were only interested in creating a great product. In the end, some of the legacy of what we did was to expose the hardware (HW) vs software (SW) tradeoffs, which resulted in things such as more sophisticated toolchains. The interesting part for that whole generation of RISC processors (MIPS, SPARC, Alpha, POWER, etc.) is how they were used in everything from automotive to space exploration. I can’t predict all the places RISC-V will be used or what technology we develop today will be viewed as world changing when we look back.
So we too are just trying to create a great architecture. It will be used in many obvious and novel ways. I am proud and amazed at the incredible designs and projects the community is working on.
Keeping “Black Mirror” in mind can you see any potential drawbacks about this technology that people should think more deeply about?
I am a change monger, I love change. The one thing I am sure of is that change will happen. It doesn’t matter what aspect of society that you look at, whether non-tech or hi-tech, it all has the potential to be used for nefarious ends. In my opinion, our jobs as humans is to make sure there is a balance, and that the net is positive and constructive. The open source technology community is often above that, wanting to break down barriers and boundaries, spark innovation and optimize our efforts. I am hopeful that our trajectory is up and to the right.
Was there a “tipping point” that guided you to open source technology? Can you tell us that story?
I have been involved with open source for a long time. I worked with GNU EMACS in the 1980s. I brought Linux compatibility to Solaris, open sourced public key infrastructure (PKI), and funded the open source Berkeley Software Distribution (BSD) version of NFSV4, and I began the arduous task of open sourcing Solaris.
Eventually I ended up at many companies using Linux and other open source products. It was clear to me that the economic models around proprietary SW were not sustainable.
People could lose their jobs in the 1990s picking Linux. Now it is a no-brainer. It only made sense that the technology would find its way into instruction set architectures (ISAs) and chip designs. It began with existing ISAs, like SPARC and POWER, donating its technology as a whole. However, those chips still require their originators to support the bulk of the effort. RISC-V is the first ISA of this magnitude to be born and developed in open source just like Linux. There is a pride of ownership with Linux that you experience with the members involved in designing RISC-V. While today we are an emerging technology, like Linux, we will be a no-brainer for designers in the future. So that’s a long winded answer to say I like the open source movement and I think it is inevitable. It spurs on innovation in ways we are only beginning to imagine.
What do you need to lead this technology to widespread adoption?
We need the successful proliferation of products using RISC-V. This will stem from finishing the things we have started and identifying and addressing gaps that are needed to meet a broad set of industry necessities.
What have you been doing to share this idea? Have you been using any innovative marketing or teaching strategies?
This answer may be best left to my marketing colleagues to answer. I try to lead every effort in an open, communicative, technologically enabling, and collaborative way. I represent and evangelize RISC-V with every interaction I have.
None of us are able to achieve success without some help along the way. Is there a particular person who you are grateful towards who helped get you to where you are? Can you share a story about that?
There is not one, but two I’d like to mention: Larry Weber and Anil Gadre. Larry was my first boss at MIPS and then I worked for him twice more at other companies. He taught me planning and how to get from point A to point B without drama. Anil was my boss at Solaris and enabled and helped me grow into my first big job. He helped me accomplish an amazing amount in an environment with many conflicting priorities. Both of these leaders enabled me to succeed by being considerate, while also challenging me. I already feel the same way about my boss at RISC-V, Calista Redmond. She is an extraordinary leader.
How have you used your success to bring goodness to the world?
Some may answer this question and the following questions in terms of technology. I will not. I don’t keep this a secret but I also haven’t shared this (up until now) with such a broad audience. For me kindness is paramount. After nine years of marriage, my wife died of cancer in 2016, culminating a long battle with the disease. The events of our lives change us. I try to both spend time volunteering in person (before COVID) and contributing money to enhance the quality of life (at least a little) for patients while they are receiving chemotherapy treatment. In my opinion, we must all be vigilant in all aspects of our lives to bring goodness to the world.
What are your “5 Things I Wish Someone Told Me Before I Started” and why. (Please share a story or example for each.)
Truthfully nothing. I have done this long enough that I just look at problems as things to be solved. How can we, collectively, get from point A to point B? I am very rarely surprised or upset (I am human and have my moments like anyone else). When there are challenges, I have Calista and our small team, as well as the RISC-V members to brainstorm with.
You are a person of great influence. If you could inspire a movement that would bring the most amount of good to the most amount of people, what would that be?
I would want to inspire everyone to always act with kindness.
Can you please give us your favorite “Life Lesson Quote”? Can you share how that was relevant to you in your life?
I am a Voltaire fan, so this quote from “Candide” really resonates with me: “Tend your garden.” If we all just did the work before us and accomplished even small things in our little piece of the world, we would feel good from that accomplishment and others would also benefit from it. Accomplishment makes for happy technologists, happy people, and a happier world. If everyone did the same, imagine what a world this would be.
Some very well-known venture capitalists (VCs) read this column. If you had 60 seconds to make a pitch to a VC, what would you say?
I have done that before and your question makes me laugh. Within the context of RISC-V, I would say: We stand on the shoulders of those who came before us including the open source SW folks and those who created the computers that help us run this amazing technology-based world. We have this great opportunity to leverage that history. We have the opportunity to leverage our open source HW community. We should spend our time innovating and not duplicating. Thirty years ago it would have taken 200 million dollars, 200 people and four years to do XYZ, and now with RISC-V companies can innovate with a smaller amount of people and dollars, as well as a shorter time to market. That means a better return on investment (ROI), better products, and better investment.
How can our readers follow you on social media?