Welcome!

Microservices Expo Authors: Pat Romanski, Flint Brenton, Elizabeth White, Liz McMillan, Jason Bloomberg

Related Topics: Containers Expo Blog, Microservices Expo, @CloudExpo, Cloud Security, @DXWorldExpo, SDN Journal

Containers Expo Blog: Article

Bare Metal Blog: Quality Is Systemic, or It Is Not

In all critical systems the failure of even one piece can have catastrophic results for the user

February 5, 2013

BareMetalBlog talking about quality testing of hardware, in all its forms. F5 does a great job in this space.

For those of you new to the Bare Metal Blog series, find them all right here.

In all critical systems – from home heating units to military firearms – the failure of even one piece can have catastrophic results for the user. While it is unlikely that the failure of an ADC is going to be quite so catastrophic, it can certainly make IT staff’s day(s) terrible and cost the organization a fortune in lost revenue. That’s not to mention the problems that downtime’s impact on an organizations’ brand can have over the longer term. It is actually pretty scary to ponder the loss of any core system, but one that acts as a gateway and scaling factor for remote employee workload and/or customer access is even higher on the list of Things To Be Avoided ™.

In general, if you think about it the number of hardware failures out there is relatively minimal. There are a ton of pieces of network gear doing their thing every day, and yes, there is the occasional outage, but if you consider the number of devices NOT going down on a given day, the failure rate is very tiny.

Still, no one wants to be in that tiny percentage any more than they absolutely must. Hardware breaks, and will always do so, it is the nature of electronic and mechanical things. But we should ask more questions of our vendors to make certain they’re doing all that they can to keep the chances of their device breaking during their otherwise useful lifetime to a minimum.

For an example of doing it right, we’ll talk a bit about the lengths that F5 goes to in an attempt to make devices as reliable as possible from an  electro-mechanical perspective. While I am an F5 employee, I will note that there is no doubt that F5 gear is highly reliable. It was known for quality before I came to F5, and I have not heard anything since joining that would change that impression. So I use F5 because (a) I am aware of the steps we take as an organization and (b) because our hardware testing is an example of doing it right.

And of course, there are things I can’t tell you, and things that we just will not have room to delve into very deeply in this overview blog. I am considering extending the Bare Metal Blog series to include (among other things) more detail about those parts that I would want to know more about if I were a reader, but for this blog, we’re going to skim so there is space to cover everything without making the blog so long you don’t read to the end.

I admit it, I’ve talked to a lot of companies about testing over the years, and can’t recall a vendor that did a more thorough job – though I can think of a few whose record in the field says they probably have a similar program. So let’s look at some of the quality testing done on hardware.

Parts are not just parts.
An ADC, like any computerized system, is a complex beast. There is a lot going on and the quality of the weakest link is the piece that sets the life expectancy and out-of-the-box quality standards for the overall product. As such there are some detailed parts and subassembly tests that gear must go through.

For F5, these tests include:

  • Signal Integrity Tests to test for signal degradation between parts/subsystems.
  • BIOS Test Suites to validate that BIOS performs as expected and handles exception cases reliably.
  • Software Design Verification Testing to detect and eliminate software quality issues early in the development process.
  • Sub- Assembly Tests to verify correct subsystem performance and quality.
  • FPGA System Validation Tests determines that the FPGA design and hardware perform as expected.
  • Automated Optical Inspection used on the PCB production line to prevent and detect defects.
  • Automated X-Ray Inspection takes 3D slices of an assembled circuit board to prevent and detect defects.
  • In-Circuit Test using a series of probes to test the populated circuit board with power applied to detect defects.
  • Flying Probe uses a “golden board” (perfect sample) to compare against a newly produced board to verify there are no defects.

Now that’s a lot of testing, though I have to admit I’m still learning about the testing process, there may well be more. But you’ll note that some things aren’t immediately called out here – like items picked from suppliers, which could be caught in some of these tests but might not  either. That is because supplier quality standards are separate from actual testing, and require that suppliers whose parts make it into F5 gear are up to standard.

Supply demands
So what do we, as an organization, require from a quality perspective of those who wish to be our suppliers? Here’s a list. This list I KNOW isn’t complete, because I pared it down for the purposes of this blog. I think you’ll get the idea from what’s here though.

  • All assembly suppliers are ISO9000 and 140001 certified.
  • Suppliers assemble and test their products to F5 specifications.
  • Suppliers are monitored with closed loop performance metrics including delivery and quality.
  • Formal Supplier Corrective Action Response program – when a fault is determined in supplier quality, a formal system to quickly address the issue.
  • Quarterly reviews with senior management utilizing a formal supplier scorecard to evaluate supplier quality, stability, and more.

The biggest one in the list, IMO, is that suppliers assemble and test product to F5 specifications. Their part is going in our box, but our name is going on it. F5 has a vested interest in protecting that name, so setting the standards by which the suppliers put together and test the product they are supplying is huge. After all, many suppliers are building tiny little subsystems for inside an F5 device, so holding them to F5 standards makes the whole stronger.

By way of example, we require the more reliable but more expensive version of capacitors from our suppliers. For a bit of background on the problem, there is an excellent article on hardwaresecrets.com (and a pretty good overview on wikipedia.com) about capacitors. By demanding that our suppliers use better quality components, the overall life expectancy of our hardware is higher, meaning you get less calls in the middle of the night.

The whole is different than the sum of the parts
While an organization can test parts until the sun rises in the west, that will not guarantee the quality of the overall product. And in the end, it is the overall product that a vendor sells. As such, manufacturers generally (and F5 specifically) keep an entire suite of whole-product tests on-hand for product quality assessment. Here are some of them used at F5.

  • Mechanical Testing Test the construction of the system by  applying shock, drop, vibe, repetitive insertion/extractions, and more.
  • Highly Accelerated Life Testing -  Heat and vibration are used to determine the quality and operational limits of the device. The goal is to simulate years of use in a manageable timeframe.
  • Environmental Stress Screening – Expose the device to extremes of environment, from temperature to voltage.
  • MFG Test Suite System Stress testing - turn everything on, Reboot, Power Cycle, et cetera. By way of example, we cycle power up to 10,000 times during this testing.
  • On-Going Reliability Testing - The products currently in the manufacturing line are randomly picked and then put in a burn-in chamber which then test the device at elevated temperature.
  • Post Pack out Audit – Pull random samples from our finished good inventory to verify quality.

That’s a lot of testing, and it is not anywhere near all that F5 does to validate a box. For example, while software testing got a hat-tip at the component level, our Traffic Management Operating System (TMOS) has a completely separate set of testing, validation, and QA processes that are not listed here because this is the Bare Metal Blog. Maybe at some point in the future I’ll do a series like Bare Metal Blog on our software. That would be interesting for me, hopefully for you also.

It’s not over when it’s over
The entire time that Lori and I were application developers, there was a party to celebrate every time we finished a major piece of software. From an evening out with the team when our tax prep software shipped to a bottle of champagne on the roof of an AutoDesk office building when AutoCAD Map shipped, we always got to relax and enjoy it a bit.

While our hardware dev teams get something similar, our hardware test teams don’t pack up the gear and call it a product. For the entire lifecycle of an F5 box – from first prototype to End of Life – our test team does continuous testing to monitor and improve the quality of the product. Unlike most of what you will find in this blog, that is pretty unique to F5. Other companies do it, but unlike ISO certification or HALT testing, continuous testing is not accepted as a mandatory part of product engineering in the computing space. F5 does this because it makes the most sense. From variations in quality of chips to suppliers changing their suppliers, things change over the production of a product, and F5 feels it is important to overall quality to stay on top of that fact. This system also allows for continuous improvement of the product over its lifecycle.

One of the many reasons I think F5 is a great company. I have twice run into scenarios that involved a vendor who did not do this type of testing, and it cost me. Once was as a reviewer, which means it was worse for the vendor than for me, and once as an IT manager, which means it was worse for me than the vendor. I would suggest you start asking your vendors about lifetime testing, because a manufacturing or supplier change can impact the reliability of the gear. And if it does, either they catch it, or you could be walking into a nightmare. The perfect example (because so many of us had to deal with it) was a huge multinational selling systems with “DeskStar” disks that we all now lovingly call “Death Star” disks.

You can rely on it
This process is a proactive investment by F5 in your satisfaction. While you might think “doesn’t all that testing – particularly when continuous testing occurs over the breadth of devices you sell – cost a lot of money?”, the answer is “nowhere near as much as having to visit every device of model X and repair it, nowhere near as much as the loss of business persistent quality issues generates”. And it is true. We truly care about your satisfaction and the reliability of your network, but when it comes down to it, that caring is based upon enlightened self interest. The net result though is devices you can trust to just keep going.

I know, we have one in our basement from before we came to F5, It’s old and looks funny next to our shiny newer one. But it still works. It’s EOL’d, so it isn’t getting any better, and when it breaks it’s done, but the device is nearly a decade old, and still operates as originally advertised.

If only our laptops could do that.

More Stories By Don MacVittie

Don MacVittie is founder of Ingrained Technology, A technical advocacy and software development consultancy. He has experience in application development, architecture, infrastructure, technical writing,DevOps, and IT management. MacVittie holds a B.S. in Computer Science from Northern Michigan University, and an M.S. in Computer Science from Nova Southeastern University.

@MicroservicesExpo Stories
Learn how to solve the problem of keeping files in sync between multiple Docker containers. In his session at 16th Cloud Expo, Aaron Brongersma, Senior Infrastructure Engineer at Modulus, discussed using rsync, GlusterFS, EBS and Bit Torrent Sync. He broke down the tools that are needed to help create a seamless user experience. In the end, can we have an environment where we can easily move Docker containers, servers, and volumes without impacting our applications? He shared his results so yo...
Don’t go chasing waterfall … development, that is. According to a recent post by Madison Moore on Medium featuring insights from several software delivery industry leaders, waterfall is – while still popular – not the best way to win in the marketplace. With methodologies like Agile, DevOps and Continuous Delivery becoming ever more prominent over the past 15 years or so, waterfall is old news. Or, is it? Moore cites a recent study by Gartner: “According to Gartner’s IT Key Metrics Data report, ...
Kubernetes is a new and revolutionary open-sourced system for managing containers across multiple hosts in a cluster. Ansible is a simple IT automation tool for just about any requirement for reproducible environments. In his session at @DevOpsSummit at 18th Cloud Expo, Patrick Galbraith, a principal engineer at HPE, discussed how to build a fully functional Kubernetes cluster on a number of virtual machines or bare-metal hosts. Also included will be a brief demonstration of running a Galera MyS...
Enterprise architects are increasingly adopting multi-cloud strategies as they seek to utilize existing data center assets, leverage the advantages of cloud computing and avoid cloud vendor lock-in. This requires a globally aware traffic management strategy that can monitor infrastructure health across data centers and end-user experience globally, while responding to control changes and system specification at the speed of today’s DevOps teams. In his session at 20th Cloud Expo, Josh Gray, Chie...
Many organizations are now looking to DevOps maturity models to gauge their DevOps adoption and compare their maturity to their peers. However, as enterprise organizations rush to adopt DevOps, moving past experimentation to embrace it at scale, they are in danger of falling into the trap that they have fallen into time and time again. Unfortunately, we've seen this movie before, and we know how it ends: badly.
Agile has finally jumped the technology shark, expanding outside the software world. Enterprises are now increasingly adopting Agile practices across their organizations in order to successfully navigate the disruptive waters that threaten to drown them. In our quest for establishing change as a core competency in our organizations, this business-centric notion of Agile is an essential component of Agile Digital Transformation. In the years since the publication of the Agile Manifesto, the conn...
"I focus on what we are calling CAST Highlight, which is our SaaS application portfolio analysis tool. It is an extremely lightweight tool that can integrate with pretty much any build process right now," explained Andrew Siegmund, Application Migration Specialist for CAST, in this SYS-CON.tv interview at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.
In IT, we sometimes coin terms for things before we know exactly what they are and how they’ll be used. The resulting terms may capture a common set of aspirations and goals – as “cloud” did broadly for on-demand, self-service, and flexible computing. But such a term can also lump together diverse and even competing practices, technologies, and priorities to the point where important distinctions are glossed over and lost.
"I will be talking about ChatOps and ChatOps as a way to solve some problems in the DevOps space," explained Himanshu Chhetri, CTO of Addteq, in this SYS-CON.tv interview at @DevOpsSummit at 20th Cloud Expo, held June 6-8, 2017, at the Javits Center in New York City, NY.
For organizations that have amassed large sums of software complexity, taking a microservices approach is the first step toward DevOps and continuous improvement / development. Integrating system-level analysis with microservices makes it easier to change and add functionality to applications at any time without the increase of risk. Before you start big transformation projects or a cloud migration, make sure these changes won’t take down your entire organization.
The Jevons Paradox suggests that when technological advances increase efficiency of a resource, it results in an overall increase in consumption. Writing on the increased use of coal as a result of technological improvements, 19th-century economist William Stanley Jevons found that these improvements led to the development of new ways to utilize coal. In his session at 19th Cloud Expo, Mark Thiele, Chief Strategy Officer for Apcera, compared the Jevons Paradox to modern-day enterprise IT, examin...
The taxi industry never saw Uber coming. Startups are a threat to incumbents like never before, and a major enabler for startups is that they are instantly “cloud ready.” If innovation moves at the pace of IT, then your company is in trouble. Why? Because your data center will not keep up with frenetic pace AWS, Microsoft and Google are rolling out new capabilities. In his session at 20th Cloud Expo, Don Browning, VP of Cloud Architecture at Turner, posited that disruption is inevitable for comp...
When you focus on a journey from up-close, you look at your own technical and cultural history and how you changed it for the benefit of the customer. This was our starting point: too many integration issues, 13 SWP days and very long cycles. It was evident that in this fast-paced industry we could no longer afford this reality. We needed something that would take us beyond reducing the development lifecycles, CI and Agile methodologies. We made a fundamental difference, even changed our culture...
High-velocity engineering teams are applying not only continuous delivery processes, but also lessons in experimentation from established leaders like Amazon, Netflix, and Facebook. These companies have made experimentation a foundation for their release processes, allowing them to try out major feature releases and redesigns within smaller groups before making them broadly available. In his session at 21st Cloud Expo, Brian Lucas, Senior Staff Engineer at Optimizely, discussed how by using ne...
In his session at 20th Cloud Expo, Mike Johnston, an infrastructure engineer at Supergiant.io, discussed how to use Kubernetes to set up a SaaS infrastructure for your business. Mike Johnston is an infrastructure engineer at Supergiant.io with over 12 years of experience designing, deploying, and maintaining server and workstation infrastructure at all scales. He has experience with brick and mortar data centers as well as cloud providers like Digital Ocean, Amazon Web Services, and Rackspace. H...
Without a clear strategy for cost control and an architecture designed with cloud services in mind, costs and operational performance can quickly get out of control. To avoid multiple architectural redesigns requires extensive thought and planning. Boundary (now part of BMC) launched a new public-facing multi-tenant high resolution monitoring service on Amazon AWS two years ago, facing challenges and learning best practices in the early days of the new service.
In his general session at 19th Cloud Expo, Manish Dixit, VP of Product and Engineering at Dice, discussed how Dice leverages data insights and tools to help both tech professionals and recruiters better understand how skills relate to each other and which skills are in high demand using interactive visualizations and salary indicator tools to maximize earning potential. Manish Dixit is VP of Product and Engineering at Dice. As the leader of the Product, Engineering and Data Sciences team at D...
You often hear the two titles of "DevOps" and "Immutable Infrastructure" used independently. In his session at DevOps Summit, John Willis, Technical Evangelist for Docker, covered the union between the two topics and why this is important. He provided an overview of Immutable Infrastructure then showed how an Immutable Continuous Delivery pipeline can be applied as a best practice for "DevOps." He ended the session with some interesting case study examples.
Both SaaS vendors and SaaS buyers are going “all-in” to hyperscale IaaS platforms such as AWS, which is disrupting the SaaS value proposition. Why should the enterprise SaaS consumer pay for the SaaS service if their data is resident in adjacent AWS S3 buckets? If both SaaS sellers and buyers are using the same cloud tools, automation and pay-per-transaction model offered by IaaS platforms, then why not host the “shrink-wrapped” software in the customers’ cloud? Further, serverless computing, cl...
Without lifecycle traceability and visibility across the tool chain, stakeholders from Planning-to-Ops have limited insight and answers to who, what, when, why and how across the DevOps lifecycle. This impacts the ability to deliver high quality software at the needed velocity to drive positive business outcomes. In his general session at @DevOpsSummit at 19th Cloud Expo, Eric Robertson, General Manager at CollabNet, will discuss how customers are able to achieve a level of transparency that e...