Mark III Systems Blog

Mark III & Watson Health: Partnering in the Cognitive Era!


Mark III is very excited to be partnering with IBM Watson Health as part of a new Watson Health partner program.  To kick off our partnership, we recently attended a briefing and collaboration session at Watson Health HQ in Cambridge, MA.  Having initially grown up in the shadows of the Texas Medical Center over the past two decades (with clients now spanning across North America), healthcare is and continues to be our longest-tenured industry and one of the industries we have the deepest skills in support of.




Accordingly, we were incredibly excited about the vision and direction of Watson Health and how it fits with our unique capabilities in support of healthcare clients.




Watson Health, as of December 2016, currently comprises of the pillars of Watson for Oncology, Watson for Genomics, Clinical Trial Matching, Care Management, Population Health Management, and Imaging (Merge), as well as pillars that revolve more around life sciences, including Watson for Drug Discovery, Watson Health Core, and Watson Health for ResearchKit.  I say December 2016 because, if the last 18 months are any guide, IBM continues to add to the portfolio and I would imagine there could be more pillars and options in the near future.

For us, one of the best parts of our visit to Cambridge was the immersion room at the Watson Health Experience Center, which weaved together many of the key Watson Health pillars with healthcare-centric cognitive approaches to show how the overall care of an actual “example” cancer patient could be possibly improved.



If you ever have a chance to attend a briefing at Watson Health HQ, we would definitely recommend it.. it’ll open up your eyes to what’s possible now and over the next 5-10 years within healthcare.  It’s not enough to simply have AI or a cognitive or deep learning system, but the need to surround it with the right types of partnerships, data, integrations, and domain expertise are critical and we love the strategy that Watson Health is employing.

With all that said, what unique complementary skills and experiences does Mark III bring to the table that can supplement the Watson Health story?

  • Developers, developers, developers:  If you’re looking to build something specific to your institution with a certain UX or engagement model, whether it be using Watson Health Core or another Watson Health focused platform, our digital development arm can assist.  We’ve been working with the IBM development cloud for over 4 years (including Watson Ecosystem) and we can help you plan for or build what you’re thinking of.  It will be fascinating to watch as more cognitive and AI-centric capabilities diffuse into the Watson Health portfolio and we can provide our perspectives and thoughts on that as it happens, including helping to build a prototype, if you’re just brainstorming.
  • Unique blend of skills to support healthcare:  Among our staff is a 10-year veteran and former Director at a Research Institute of a large healthcare provider, and a PhD in Bioinformatics, along with our team of developers, architects, and engineers.  We pride ourselves on our “full stack” approach that’s aligned with what institutions are trying to achieve in the digital and cognitive era (but with knowledge of the past and present), and this is just one example.
  • Optimization and enablement of existing core tech stacks: Most of these tech stacks are key current application sets (EMR/EHR, Imaging, ERP, Research, etc.) that currently run on-premise in a datacenter.  Since these key systems feed Watson, it’s important that their foundation is strong, which includes anything from servers to storage to networking to virtualization at any layer to software tooling to middleware.  This is an area where our team of systems engineers and architects can definitely help ensure that your current strategy is strong so that you can capitalize on a future enabled by Watson Health.
  • Integrations:  Watson Health and hybrid digital strategies, in general, rely upon the ability to integrate together services and platforms in an open, yet secure, manner, which requires a blend of all the capabilities listed above.  Whether the need is to tie together APIs, connect the datacenter to other datacenters or the IBM Cloud, build your own cognitive platform, or optimize/refresh the IT infrastructure foundation for your EMR/EHR or Research/HPC, we have the unique capabilities and history to enable this.


Stay tuned!  We’re incredibly excited about the future and I know we’ll have some more exciting concepts and news to share in the near future around Watson Health.




Posted in Bluemix, Cloud Object Storage, Cognitive Computing, IBM Watson, watson, watson health | Tagged , , , , , , , , , , , , , , , , , | Leave a comment

IBM Software-Defined Storage & Video: Building the Foundation for Video Analytics & Computer Vision


Video is one of the fastest growing forms of data in the digital world today, especially in the enterprise.  According to industry forecasts, video as a percentage of Internet traffic is projected to rise above 80% in the coming years, with a large portion of that volume connected to the enterprise and how enterprises engage with their major stakeholders.

Here at Mark III we’re fortunate to partner with IBM on not only cost-effectively and flexibly addressing the video storage challenges of today, but also leaning on these foundational video storage strategies of today to be able to fully take advantage of the emerging capabilities of tomorrow focused around video analytics and AI-powered computer vision that will revolutionize industries over the next 10 years.  In short, the decisions made around how to best store video data in 2016 and 2017 will impact how your enterprise will be able to take advantage (or not take advantage) of these huge digital competitive disruptors in future years.

We think that as AI-powered video analytics capabilities develop further, the volume and need for video will explode even more, as video will not only be a type of media used for historical recording and engagement with customers/employees/partners, but also one that can quickly be used in real-time to enable an enterprise to augment their business decision-making with ultra-responsive, human-like judgement and advice (all powered by video).  The video itself could possibly be employed by any number of IoT devices, cameras, or even bots.

As for the right now, we believe a foundation of IBM Cloud Object Storage or a platform in the IBM Spectrum Storage family can enable this future, depending on the specific requirements.  Specifically, we’re incredibly excited about IBM Cloud Object Storage and its capability of enabling a cost-effective, flexible, and secure platform for unstructured data, including video, that can be served out via open RESTful APIs.  A leader in the object storage market since its founding in 2004, IBM Cloud Object Storage (formerly Cleversafe) has been rated by Gartner as the leader for Overall Use Case in its “Critical Capabilities for Object Storage” study in March 2016.  IBM Cloud Object Storage, which can scale to 100s of petabytes (and beyond), can be deployed in an all on-premise, all in the IBM Cloud, or in a hybrid model between local datacenters and the IBM Cloud.




As for a glimpse of the future, Mark III’s digital development arm, BlueChasm, took the IBM Cloud Object Storage foundation (deployed on-premise on IBM Systems in our lab for our prototype) and created its own video analytics platform (called VideoRecon), which uses IBM Watson to “watch” and “listen” to videos and auto-tag the videos with metadata of what it “sees” and “hears.”  IBM Cloud Object Storage serves as the flexible object storage layer for the videos and tags, and its open API access enables other platforms or applications now and in the future to access its repository in a frictionless manner.  Here is the original video analytics prototype that BlueChasm built early in 2016, which served as the original foundation for what VideoRecon has grown to today.

VideoRecon is just one possible platform built on the IBM stack that can take advantage of IBM Cloud Object Storage for video insights.  Since IBM Cloud Object Storage is driven as an open API, you actually don’t even need to use the IBM Cloud stack to tie into it or use it for your particular use case…  the idea is that it can serve as the digital data foundation for virtually any use case and any stack you can think of that incorporates video.

If you need help bringing IBM Cloud Object Storage to life, Mark III can definitely apply our “full stack” approach and skills to assist, including not only our engineers that are skilled in the architecture and deployment of IBM Cloud Object Storage (or IBM Spectrum Storage and/or Systems), but also our developers and DevOps teams that are available to help you build platform and app prototypes on top of IBM Cloud Object Storage or just advise you, in general, on how to get the most out of the power of the IBM Cloud (Bluemix/Watson/Cloudant/etc).




Posted in Bluemix, Cleversafe, Cloud Object Storage, Cognitive Computing, IBM Watson, object storage, Spectrum Storage, storage, Uncategorized | Tagged , , , , , , , , , , , , , , , , , , | Leave a comment

IBM Cloud Object Storage: The foundation of the digital and cognitive eras!


As a long-time IBM Premier Business Partner, we’re very excited about IBM’s announcement today that IBM Cloud Object Storage, based on Cleversafe, is now supported across 40 IBM Cloud datacenters globally, which gives enterprises and service providers a perfect public and hybrid cloud option to complement Cloud Object Storage’s long-time, industry-leading on-premise object storage capabilities.

Rated by Gartner as the leader for Overall Use Case in its “Critical Capabilities for Object Storage” study in March 2016, we view IBM Cloud Object Storage as the key to not only solving immediate unstructured data challenges, but also serving as the digital foundation to enabling key enterprise initiatives of the digital era, including IoT, cloud, cognitive, deep learning, social, mobile, and next-generation analytics.

At Mark III, we’ve had IBM Cloud Object Storage up and running in our lab since the Cleversafe acquisition was announced by IBM in 2015, and have worked hand in hand with our clients on use cases ranging from video data to backups to archival data to digital content repository data and many more, since that time.  In short, if you need to store unstructured data in a cost-effective and flexible manner with the possibility of scaling to 100s of PBs (most likely spanning multiple sites) and be positioned for future digital initiatives, IBM Cloud Object Storage represents a truly unique platform to enable this, all with simple access via popular object storage APIs (S3, OpenStack Swift, etc).




Here’s a look at our very first IBM Cloud Object Storage (Cleversafe) dsNet instance we spun up last year in our demo center/lab, just after the acquisition was announced by IBM:



We’re huge fans of IBM Cloud Object Storage partly because of the versatility in how you can set it up.  Cleversafe, as a company, was founded in 2004 first as an on-premise, object storage platform years BEFORE any of the mainstream public cloud object storage platforms existed.  As a result of this head start, Cleversafe was able to build a lead in the industry by loading up on advanced features and offering rock-solid stability, due to its vast market adoption and customer field cycles.

With this announcement of IBM Cloud Object Storage in the IBM Cloud, you can now easily customize and build a Cloud Object Storage platform based on 100% on-premise, 100% public cloud, or a hybrid model unique to your enterprise.  For any on-premise component, you can choose if you’d like to use Cloud Object Storage appliances, or use just the Cloud Object Storage software on top of general-purpose x86 servers.

In addition to helping our clients with the architecture and rollout of IBM Cloud Object Storage for their own use cases, we’ve also worked closely with our own BlueChasm digital development unit to develop prototype use cases that leverage Cloud Object Storage as the object storage layer (as shown below).  Additionally, we’ve also contributed to open source to enable developers anywhere to more easily create apps that can leverage IBM Cloud Object Storage (npm package here).

To get your imagination going on how this might apply to your own digital initiatives, here are just some of the platform/app concepts that we’ve built on top of IBM Cloud Object Storage:

  • VideoRecon:  A video analytics platform that uses deep learning to analyze the visual and audio components of a video and return auto-tags.  The videos and tags themselves are stored on IBM Cloud Object Storage.
  • Cognitive Advertising Display:  Dubbed the “Minority Report board,” this digital advertising display looks at the person in front of the board and shows them a targeted advertisement based on their age/gender.  Ad content and video/images of subjects are stored on IBM Cloud Object Storage, and served via an advertising platform that we built
  • Cognitive Drone Platform:  Built a drone that uses IBM Watson to visually identify objects while in flight to “observe” things of interest to an enterprise (i.e. inventory levels in a warehouse or noting the composition of a crowd of customers).  IBM Cloud Object Storage stores the images used for analysis by our platform.


Regardless of if you’re looking at IBM Cloud Object Storage from the unstructured data storage angle, analytics angle, or the digital initiatives angle, we’re incredibly excited about the future of IBM Cloud Object Storage.  Let us know if you’d like to learn more about IBM Cloud Object Storage or dive into one of our digital prototypes!


Posted in Uncategorized | Tagged , , , , , , , , , , , , , , , , , , , , , | Leave a comment

Brocade Workflow Composer & StackStorm: Our contribution to the future of DevOps and ChatOps!


Through our partnership with Brocade as a Brocade Elite Partner, Mark III recently started offering solutions and capabilities built around Brocade Workflow Composer (BWC), which is a powerful enterprise-level, open automation framework that leverages the StackStorm open source project at its core.  Brocade actually acquired StackStorm earlier this year and has leaned on the StackStorm open source project as the core for BWC, while allowing the open source project itself to continue to thrive independently.

In a nutshell, StackStorm is an open, event-driven automated framework that essentially “glues together” all your enterprise apps, services, and workflows (including things like Chef, Puppet, Ansible, GitHub, Docker, VMware, IBM SoftLayer, Windows, Jenkins, Slack, Twitter, Splunk, Twilio, Cassandra, Elasticsearch, ServiceNow, and many more) .  In short, you can have events in one of your tech “stacks” trigger events or actions in others, in a manner that requires virtually no setup (with nearly 2000 pre-wired integrations from the project), and done in an extendable and flexible manner with a focus on DevOps and ChatOps.



Personally, I like to think of StackStorm (Brocade Workflow Composer) as the “automation engine of automation engines” (especially if you’re already using multiple configuration management tools across your operation).  And even better, with the addition of StackStorm to the Brocade family, StackStorm and Brocade Workflow Composer now also support Brocade’s networking portfolio, which means that all of StackStorm’s DevOps and ChatOps integrations can now be extended to the networking stack too!

Use cases?  Auto-remediation, auto-scaling, ChatOps (two-way!), repeatable/repetitive workflows (pre-planned or event-based), and much more!

With all that said, due to some feedback from our ecosystem partners, we decided to actually write and contribute an integration pack to the StackStorm project itself, specifically one for Zendesk, which is the one of the most popular customer service software and support ticketing platforms in use today.  We actually use Zendesk ourselves internally, so this worked out for us personally as well.

The Zendesk integration pack we put together (found here) would enable any user of StackStorm or Brocade Workflow Composer to be able to do the following in Zendesk:

  • Create tickets
  • Update tickets
  • Change the status of tickets
  • Close tickets
  • Search for tickets using keywords
  • Use ChatOps to interact with Zendesk




This might mean that an enterprise (or service provider) could easily wire together an auto-remediation workflow and as part of the workflow, they could now open up a services/support ticket via Zendesk to alert the team, in the event of an issue requiring a ticket (either internally or from a third-party).

Or that a DevOps or support engineer team could easily create/update/query tickets transparently on Zendesk all without leaving their team’s Slack channel.  Pretty powerful stuff if you’re looking to get things done really quickly (for YOUR end user) and with the maximum amount of team collaboration and transparency.




If you’re looking for more information on how we built our Zendesk integration pack, I’d recommend you check out the BlueChasm blog here (BlueChasm is a our digital development team).

Let us know if you have any questions about Brocade Workflow Composer, StackStorm, or Brocade (in general)– the future of DevOps in our part of the world looks bright!



Posted in Uncategorized | Tagged , , , , , , , | Leave a comment

Barreleye is here . . . General availability!


Today at IBM Edge 2016, Mark III and our partners in the OpenPOWER Foundation are announcing the immediate availability of an OpenPOWER server platform based on the Barreleye Open Compute Project (OCP) design.

We’re doing this announcement specifically in partnership with Penguin Computing under the OCP-compatible model of the Penguin Magna 1015, which provides an enterprise supported version of the Barreleye system.  As a long-time IBM Premier Business Partner with two decades of experience with POWER, our strong team of engineers are also available to offer their expertise and services around the Magna 1015 platform to ensure that our joint OpenPOWER clients are successful.



If you might recall, Barreleye is based on the Rackspace-led OCP design that incorporates OpenPOWER technologies (including POWER8 processors), and was a system that Mark III announced back in March at the OCP Summit that it would be offering very soon.

We view the Magna 1015 (Barreleye) as fitting a key niche in our portfolio of OpenPOWER platforms, as many hyperscale users of compute have looked at or are starting to look at OCP approaches to maximizing datacenter efficiency as they grow.

As a member of both foundations, we’re very excited about the future of both OpenPOWER and OCP in delivering highly efficient architectures for the bandwidth-intensive workloads of the next decade.  To us, Barreleye is the culmination of both these industry movements, but is also just the beginning of a new wave of innovation.




Posted in Uncategorized | Leave a comment

Denser than DISK: The NEW IBM DeepFlash 150 takes flash storage to new levels


512TB of FLASH in 3U?  7 petabytes of flash in one rack?  Less than $1/GB?  Really?

With the announcement of the new IBM DeepFlash 150, IBM has brought forth a uniquely designed flash storage system targeted at cost-effective, high-density flash deployments for unstructured data use cases, including Big Data Analytics and HPC.  Highly complementary to the existing IBM FlashSystem line (including models like the new FlashSystem A9000), which are more focused around database-driven, OLTP, and virtualization use cases, the DeepFlash line will enable clients to achieve incredible performance improvements at a cost-effective price points for unstructured data use cases (which make up around 80% of enterprise data today).



Combined with IBM’s award-winning Spectrum Scale software, the DeepFlash 150 is an incredible Exabyte-scale, high-density platform to enable data storage stacks for use cases like Hadoop/Spark, NoSQL databases, HPC, Research, and OpenStack, for both file and object-based storage accessibility.




Although the DeepFlash 150 does not necessarily NEED to be attached to Spectrum Scale, there are a lot of unique benefits that come with its attachment (as this next slide points out):




From a Mark III perspective, we see some exciting potential with this system, as it relates to not only HPC and Research, but also supporting the huge data needs of the fast emerging wave of digital applications, most of which use “shared nothing” compute architectures within which we strongly feel that the DeepFlash 150 will excel in.

Whether you’re looking for guidance at the system/storage/networking level or higher in the stack from a DevOps or developer/design perspective, Mark III is ready to assist.  Feel free to reach out via chat in the bottom right hand corner if you’d like to continue the conversation!


Posted in efficiency, flash, FlashSystem, hadoop, HPC, IBM Watson, openstack, storage, Uncategorized | Tagged , , , , , , , , , , , , , , | Leave a comment

Our Recap: Dev Day Houston 2016 with Mark III!


Mark III Systems, along with some of our key technology ecosystem partners, hosted the first ever Dev Day Houston meetup and expo last week at Karbach Brewing Company in Houston.

Described as a “meetup and expo for ALL Houston tech professionals that build amazing software for enterprises, organizations, service providers, research institutions, and everything in between,” Dev Day was both an incredible exchange of ideas on WHAT to build and HOW to build software, in addition to being an awesome meeting of the Houston enterprise developer community.


DevDay-1 DevDay-2 DevDay-3 DevDay-4


In case you missed it (or if you weren’t able to get to all the tables), here’s an in-depth recap at some of the tech tables we had at Dev Day:


IBM Watson

This table had a cognitive drone that was built by our BlueChasm dev team focused on being able to “see” using an onboard camera and being able to store images and analyze those images using IBM Cloud Object Storage (Cleversafe) and IBM Watson, respectively.  All of this intelligence is powered by our open platform written on top of IBM Bluemix.

Enterprise use cases?  Any scenario which requires visual recognition of written content/people/objects at a distance that’s “costly” for businesses to traverse (in terms of either manpower or time) or into a location that’s considered dangerous (i.e. heights/proximity to power lines/chemical leaks/etc).  Our favorite example is the usage of drones to visually verify/analyze inventory in a very large warehouse, although the most popular examples are probably physical security systems or damage assessments in hard to reach places in the insurance industry.

We originally had planned on trying to fly the drone at the venue (we stuck with just showing the visual recognition in action), but with the low ceilings and ceiling fans in the venue, we don’t think our friends at Karbach would’ve liked that very much (one might say our coding skills are better than our piloting skills 🙂 )


Drone-2  Drone-table-2


In addition to the drone, we had our Video Recon video analytics platform on display.  For this event format, we used our iPad and an iOS app we designed to showcase the functionality.  For those not aware, Video Recon is a video analytics platform that our team put together that uses IBM Watson, Bluemix, and IBM Cloud Object Storage (Cleversafe), in combination with some of our own algorithms, to enable a user to leverage Watson to auto-tag a video with what Watson “sees” in the video.  These tags will allow the video to be searched, filtered, and analyzed (after tagging), all without any form of manual tagging or labeling that’s required with most video/audio storage software platforms today.




BlueChasm (Mark III’s digital development arm) had a table that featured a couple different cognitive demo platforms also built around IBM Watson, Bluemix, and IBM Cloud Object Storage.  The first was a “Suggest-a-Song” digital app that uses IBM Watson to analyze the lyrics of a song you might like and suggest another song with similar lyrics based on an algorithm.

BlueChasm-1-devday BlueChasm-2-devday


The second demo we had on hand was a real-time demonstration of the “Visual Communications” platform that our team built, also with IBM Watson, Bluemix, and IBM Cloud Object Storage.  If you aren’t familiar with the Visual Communications platform, it uses a camera and IBM Watson to “look” at the viewer of a screen, and based on the viewer’s age and gender, it shows them some form of highly targeted pre-loaded content (which can be a video, image, or audio recording).  In the past, we’ve typically demonstrated this platform with the enterprise use case of viewer-specific advertising (i.e. showing an advertisement for teen apparel to someone in the 13-18 year old demographic), but for the purposes of Dev Day, we decided to play hit songs that were popular when that viewer was a teenager.  Love the idea of using it for viewer-centric maps/directions too (thanks to our fellow Dev Day’ers that suggested that use case!).



The Weather Company

The Weather Company, a recently acquired business of IBM, also had a demo table covering their portfolio of weather data solutions, including a huge amount of weather data available via APIs (our favorite!), which can be used for specific enterprise use cases across any industry that is impacted by weather.  The geographic granularity of the weather data (down to 0.1 miles) available for our apps/platforms and our enterprise clients is truly amazing and will enable us all to do some never done before things involving the predictive power of weather forecasting and how it impacts business.


Weather Company-1


IBM Bluemix

IBM Bluemix is the cloud PaaS platform that is the core of most all of the digital apps and services that we create, and the IBM Bluemix development team was nice enough to come down from Austin and volunteer to man that particular booth at Dev Day.  Bluemix is available via Public Cloud, Dedicated (public cloud), and Local (on-premise) deployment models and can be used to build pretty much any web or mobile app/platform/service needed.  Here was the signage from Dev Day on Bluemix, where a lot of the interest was on creating and maintaining mobile apps, specifically (although Bluemix can handle all forms of digital apps, including web/mobile/emerging).





Docker and OpenStack

Mark III is a Docker partner and also offers numerous systems/technologies that utilize OpenStack for cloud automation (some which incorporate Docker as well).  As part of our Docker/OpenStack table, we had one of our DevOps engineers showcase the new Docker Datacenter suite (which we have running in our Houston lab), which includes key enterprise-centric components like Docker Universal Control Plane (UCP) and Docker Trusted Registry.  The recent Docker 1.12 release gave us some exciting things to talk to visitors about (Swarm mode!), as well as a demonstration in-person of how an enterprise-centric Containers-as-a-Service (Caas) model would work from a Docker standpoint.


Docker-1 Docker UCP BPIC


IBM Cloud Object Storage (formerly Cleversafe)

The newly renamed IBM Cloud Object Storage (aka Cleversafe) has been a focus area for both Mark III and BlueChasm over the past 8 months, since the acquisition by IBM was announced.  It’s served as the API-driven “data foundation” not only for existing applications, but also as the key storage piece for pretty much all our digital apps/platforms that we’re creating under the BlueChasm umbrella (and what we would recommend for our clients currently tackling similar digital initiatives).  This includes most of the applications we actually demonstrated at Dev Day, which all use IBM Cloud Object Storage (Video Recon, Cognitive Drone, Visual Communications, etc).

Named a leader by both Gartner and IDC and really the original pioneer of the object storage space since its founding in 2004, IBM Cloud Object Storage is a pretty amazing story in how it uniquely stores unstructured data, in and of itself.  Thanks to our local IBM Cloud Object Storage team for volunteering their time at the table!


CloudObjectStorage-1  IBMCOS-1


IBM BigInsights

Our resident Ph.D and data scientist (and systems engineer!) manned our IBM BigInsights booth (naturally).  BigInsights is IBM’s open distribution of Hadoop and Spark.




Brocade and StackStorm

Last, but certainly not least, our friends at Brocade were on hand to demonstrate StackStorm, a DevOps-friendly automation framework that Brocade recently acquired.

Using StackStorm, you can orchestrate literally any type of event-driven automation across almost any tech stacks that you can think of (AWS, GitHub, Slack, Jenkins, JIRA, Azure, OpenStack, NodeJS, etc).  The integrations are what make this platform amazing in our opinion and the creative way in which Brocade is also tying in the StackStorm automation framework (which is open source) into networking infrastructure is pretty awesome, in our opinion.  Big fan of ChatOps too… the things you can automate and do directly from Slack are amazing.




Til next Dev Day…

Thank you to everyone who attended Dev Day and for everyone who helped set up and make the meetup a huge success!

We’ll definitely be looking forward to sponsoring future editions of Dev Day not only in Houston, but other major cities around the US (Dallas and Austin, you’re next!).

If any of these technologies and/or projects/products interest you, please feel free to reach out to us directly at Mark III to start a conversation!

Our developers, DevOps engineers, Systems Engineers, and sales reps are ready to assist you in building your tech stacks, whether it be an emerging cognitive or IoT digital use case or simply refreshing a long-time, business-critical application or infrastructure stack.

Posted in biginsights, Bluemix, brocade, Cleversafe, Cognitive Computing, docker, hadoop, HPC, IBM Watson, openstack, stackstorm, Uncategorized, VMware | Tagged , , , , , , , , , , , , , , , , , , , , , , , , , | Leave a comment

Cognitive B2I Advertising with IBM Watson, Bluemix, and IBM Cloud Object Storage


If you’ve been around enough conversations in our industry about things like machine learning, AI, cognitive, and analytics, inevitably someone will bring up the scene from the movie “The Minority Report,” where Tom Cruise is walking quickly through a retail area and cameras are scanning his face and showing him advertisements personalized for him (even calling him out by name!).  Just a matter of time until that’s a reality we always say, right?

Well, thanks to the combination of IBM Watson, Bluemix, Cloud Object Storage (formerly Cleversafe), and Spectrum Storage (and a little creative coding from our BlueChasm dev team), we can honestly say that that vision of personalized advertising has become a reality in the form of a platform that our team built that we’re calling “Visual Communications.”

Just like in the movie, our Visual Communications platform enables any enterprise to interact directly with INDIVIDUALS in a personalized manner by showing them an advertisement targeted at their demographic group (age/gender/etc) based on what the screen-mounted camera detects from “looking” at their face.

The B2I (Business-to-Individual) platform uses a unique combination of cognitive technologies from the IBM Cloud tech stack.  What’s intriguing about this new B2I model (which is the term that was used recently by IBM CEO Ginni Rometty) is that its truly a step beyond B2B and even B2C, in its the ability to use cognitive capabilities to engage directly with and cater to specific individuals that might be customers or other important ecosystem partners.

This level of “on the spot” personalized engagement, whether it be targeted advertisements or other focused interactions, will enable businesses to serve their customers in extremely unique and innovative ways, especially when combined with previously gathered business data about that customer/individual.  We’ve included some possible use case suggestions in the video, but there are almost a limitless number of possibilities that build on top of our base open platform.



So how did we build this platform?

At a high-level, we used the following building blocks as part of our tech stack:

  • IBM Watson Visual Recognition:  Watson Ecosystem API that enabled some of our visual recognition capabilities in our platform
  • IBM Bluemix:  The master cloud development platform in the IBM public cloud (of which Watson is a service within)
  • IBM Cloud Object Storage (Cleversafe):  Our object storage service that our platform uses to serve up the advertisements in the form of images via its S3-compatible API (located on-premise in our lab)
  • IBM Spectrum Storage/Storwize:  Part of our IBM Cleversafe Object Storage implementation utilizes IBM Storage Systems under the covers, specifically the extremely versatile IBM Storwize line (located on-premise in our lab)
  • Python:  Our programming language of choice in this specific use case
  • BlueChasm:  Our awesome BlueChasm dev team (who actually dreamed up this use case, pulled everything together, and built the platform)


Thank you for reading this blog entry– hope you found it interesting!

Stay tuned for more things that we’re working on, and feel free to reach out to us anytime via our website or Twitter (@MarkIIISystems), if you’d like to talk more about how we built this platform or catch up on the latest on what else we have going on!


Posted in Bluemix, Cleversafe, Cognitive Computing, IBM Watson, storage, Storwize V3700, Storwize V5000, Storwize V7000 | Tagged , , , , , , , , , , , | Leave a comment

Provisioning an IBM Storwize V7000 using OpenStack Cinder and IBM Cloud Manager


The IBM Storwize V7000 is an incredibly versatile storage system that’s been highly popular with our clients over the years.  With IBM as one of the founding members of the OpenStack Foundation (in 2012), the Storwize line has featured OpenStack capabilities for many years and IBM is one of the leading contributors to the OpenStack project every year.

In lockstep, Mark III and our team of engineers has helped to support our clients that have chosen to deploy OpenStack for at least part of their environment, mostly to operate developer-centric private clouds.  The following is a very short tutorial on provisioning and mapping a volume on a Storwize V7000 in our lab via OpenStack Cinder and IBM Cloud Manager to a host that we’ll be using for Docker (for our digital development team to work from)–  the next blog entry in our series will show how we use some of Docker’s tools to manage containers from this very environment (stay tuned!).


V7000 and OpenStack Cinder Tutorial:

First, we logged into the IBM Cloud Manager dashboard and created a volume using OpenStack Cinder to be used by an instance within our Docker environment:



Second, just to verify that everything looks good, we logged onto the Storwize V7000 GUI to view the audit log.   As expected, the V7000 Cinder driver automatically connected to the V7000 and created the volume.



Next, we selected the new volume in the OpenStack dashboard on IBM Cloud Manager and attached it to one of our Docker tenant instances:



Fourth, in the previous step, we chose to attach the volume to the instance titled “Docker_Node01.”  For testing purposes, we just wanted to confirm which compute node Docker_Node01 is running on, so we queried nova services from the CLI to find out.  The output tells us which compute node it is running on:



Fifth, back in the dashboard, we can see that OpenStack Cinder attached the volume to the Docker_Node01 Ubuntu instance on /dev/vdb:



Next, the V7000 audit log shows us that OpenStack Cinder automatically detected which compute node is hosting Docker_Node01, created the host definition on the V7000 along with the node’s WWPNs and then mapped the volume to the newly created host.



Last, to verify that the volume is now mounted to the OpenStack instance, we logged into Docker_Node01 to check the output of the lsblk command.  Sure enough, there is our 10GB volume, as expected, mounted to /dev/vdb.  Success!!



We have OpenStack Cinder scenarios on various storage platforms, as well as many other OpenStack and private cloud demos running in our demo center/lab, so please feel free to reach out if there is anything you’d like to see!

Posted in openstack, storage, Storwize V7000, virtualization | Tagged , , , , , , , | Leave a comment

Dev Day Houston 2016

Dev Day banner v3

Karbach Brewery

Register Here!

Come to Dev Day Houston 2016 at the Karbach Brewery on July 14th! It is a meetup and expo for ALL Houston tech professionals that build amazing software for enterprises, organizations, service providers, research institutions, and everything in between.

Why Come to Dev Day Houston 2016?

apple watch text

Posted in Cognitive Computing, openstack | Tagged , , , , , , , | Leave a comment