I am a senior member of a team responsible for administration of a large complex GPFS (aka Spectrum Scale) multi-cluster file system configuration that provides storage for the Mira BG/Q supercomputer (#9 top 500) complex, a 60 node IBM ESS file system that is currently providing project storage and home directories to our Cray XC40 system called Theta (using native GPFS clusters and Cray's DVS IO Forwarding software). The Mira clusters are primarily based on DDN SFA12KE (embedded NSD servers) running GridScaler (RHEL). In addition to the numerous GPFS clusters in the Mira/Theta complex I also administer a Cray Sonexion 3000 (OEMed Seagate ClusterStor) supporting a 10PB Lustre file system. My duties include day to day administration tasks, software upgrades, problem determination/resolution, planning for future exploitation, writing code for automation activities as well as so many other tasks too numerous to mention in detail. This role greatly leverages my solid background both in storage systems as well as systems administration. Working at a DOE National Laboratory that is contributing to open science is really the "icing on the cake" of a really enjoyable and challenging job - one that in my time at the ALCF I am thoroughly enjoying. I'm also enjoying the opportunity and challenges associated with learning yet another parallel file system (Lustre) in depth - a learning that added to my almost 18 years of experience with GPFS gives me a very strong working knowledge of the two industry leading parallel file systems in use today.

I am contracting to IBM as a GPFS (now called Spectrum Scale) file system specialist within a team that is responsible for the development and support of high end large scale HPC computing environments running on P775 (p7IH) clusters running both AIX and Linux operating systems. These systems are using the GPFS Native Raid (GNR) capability to provide a very high I/O bandwidth, high data availability and extraordinary data reliability (exploiting features like end-to-end check-sums for example) at extreme scale. My role consists of multiple responsibilities including leading the testing of new versions of GPFS in combination with the HPC software stack in the P775 environment, acting an interface between this team and the GPFS development team where I was a team member for 14 years, and providing consulting advice to members of the extended P775 team as well as customers. This role demands various skills for it's wide range of varying challenges, scripting test cases/tools for a very specific proof, programming test suites or performance regression workloads or working on how to plan for future releases and providing an articulation of our customers needs with respect to the file system technology as a sample of tasks.I'm back to what I love to do, which is solving problems in high end HPC file system environments.

I took a personal sabbatical for the purposes of both professional and personal renewal and now that I've completed it I can say without hesitation that I achieved the goals I set for myself when starting it.During this time I augmented my object orientated programming experience in C++ by learning Java, reading various programming books including topics like OO design patterns and have emerged as a much stronger programmer for my efforts. I also explored several open source file system technologies in order to extend my file systems knowledge beyond GPFS and the typical UNIX file systems I've worked with for the last 15+ years.In January of 2015 I had the good fortune to join the local Google Development Group (GDG) here in the Hudson Valley of New York. The Hudson Valley GDG is a vibrant community of local programmers who came together under the suburb leadership of Nitya Narasimhan and Hendrix Tavarez to learn Android Programming following a curriculum co-developed by Google and Udacity. We met once a week at Dutchess Community College to have in depth discussions on the topics covered in the current Udacity course lesson. Nitya and Hendix where always on hand, both of whom have a deep level of experience in the Android programming environments to help the group understand what wasn't clear and to explain things more fully. I am now actively writing Android programs at a serious level (fragments, content providers, material design) as a hobby and really enjoying it. Android programming is quite a different domain from the programming I'd been doing as a member of a file systems development team. The GDG is alive and well and I'm looking forward to our next learning exercise be it Polymer, Meteor or whatever we decide to pursue.

I had several roles on the GPFS (aka Spectrum Scale) team over the 14 years that I was a member. First as a test team member testing the product in it's entirety, then I fulfilled the role of "device qualification"​ working with IBM and third party (EMC, HDS) device vendors to ensure disk subsystem readiness to work properly with the multi-node/multi-OS highly concurrent nature of GPFS. I was then appointed to team lead of the FVT (Functional Verification and Test) team and performed that overseeing role in addition to a the device specific tasks until I was moved to my final role on the team when I was the overall GPFS test architect. Throughout my time with GPFS I was a senior level technical skill and the "go to"​ person in many situations. As such I've had opportunities to handle customer complaints, to fly to customer locations and solve problems on-site and many other forms of dealing with problems. I've also spent time teaching GPFS on education tours to different world-wide geographies helping them to understand various aspects of the product and it's implementation. I also participated frequently as an extended member of the LSST (Large System Support Team) by performing the GPFS aspects of large system installation/customer acceptance activities at customer locations including in China, Taiwan and various countries in Europe. GPFS Test Architect - Overall responsibility to ensure complete test coverage by root cause analysis of internal/field found problems. Interface with product teams exploiting GPFS to ensure our test coverage ensured readiness for them. Developed and authored test programs.GPFS FVT Team Lead - Lead a WW team of 15-20 testing GPFS, developed and delivered education to new team members, mentored team members on test methodologies and authored test programs.GPFS Device Test Specialist/Test Generalist - Worked with various IBM and third party vendors to ensure device readiness and tested GPFS in it's entirety.

- Design and assist in implementation of solutions in a large scale multinational oil company environment, supporting both scientific and business end user customers. - Provided senior level RS/6000 and AIX support to various customer situations across the country.

- Provided AIX and SP2 technical expertise and project management skills to the first production implementation of SAP R/3 running with Oracle Parallel Server (OPS) on IBM's RS/6000 SP.- Provided technical implementation skills to Western Canadian RS/6000 and SP customers. - Contributed in a consultative role in the design of high-end client server infrastructure in the Canadian marketplace.

- Extensive involvement in bench marking in compute and I/O intensive technical computing environments -- mainframe MVS, AIX/ESA and RS/6000 AIX.- Strong experience with the design of very large scale I/O and compute server environments based on RS/6000 and AIX. - AIX specialist supporting IBM SP2 running reservoir simulation codes. - Ran a early install program for TSM/HSM (then called ADSM) in a very large UNIX (primarily SUN) environment.I was hired at SBM for my strong background in MVS systems programming. During my first year with SBM an unique opportunity was offered to learn AIX by joining the newly developed IBM UK "AIX Apprenticeship Program". During the following year I would spend alternate my time between attending hand-on AIX classes in the UK and working as an AIX specialist in Saudi Arabia honing my newly acquired skills. For many years I was the only AIX skill at SBM and as such was leaned on in various situations across the country as well as transferred my AIX skills to others in a TTT (Teach The Teacher) form.

- Responsible for MVS/ESA and related software (JES2) including performance measurement and multi-level Storage Management (HSM and DF/HSM).

- Responsible for the installation and systems programming responsibilities of a MVS/JES3 environment. - MVS was running as a guest operating system under VM so I also got a great degree of exposure to the VM operating system even though it wasn't my primary responsibility.