<iframe src="//www.googletagmanager.com/ns.html?id=GTM-WHST8N" height="0" width="0" style="display:none;visibility:hidden">

Stay Powered Blog

Marc Cram

Recent Posts

Hyperscale Data Centers: Thoughts On Support and Service?

Posted by Marc Cram on Sep 19, 2017 11:00:45 AM

businessman hand using tablet computer and server room background.jpeg

Your hyperscale data center operates on a lean budget. You want to install your hardware and be done with it until you decommission it for the next efficiency-driven replacement cycle.  But real world hardware does fail, and when it does, you want your suppliers to be both knowledgeable and responsive. They need to be able to troubleshoot remotely or on site, and get you answers and replacement product quickly so that your application can be restored.

I’m interested in knowing your thoughts:

  1. In your opinion, who offers the best technical support for PDUs?
  1. What is “state of the art” when it comes to service?
  1. How could we do a better job for you?

Email me at [email protected] or on twitter @mcram01

Your Hyperscale Data Center Demands Reliability. Why Compromise? Learn More Today

 

Topics: Service Oriented, Hyperscale

Hyperscale Data Centers: Do You Use A Bespoke Power Solution...Or Not?

Posted by Marc Cram on Aug 28, 2017 1:42:54 PM

cloud-computing-servers-customerPowerSolution_noLogo.png

You have a deadline, and you have your goals. 

Your hyperscale data center design needs to maximize power efficiency. Use free air cooling or adiabatic cooling. And support ambient air operating temperature of 25-35oC, with hot aisle exhaust temperatures approaching 60oC. 

You need lots of outlets – C13s, C19s, or even something that allows you to blind mate servers to the power strip. 

And it needs to meet regulatory requirements in most major geographical regions around the world. 

You want to keep it simple.  

And you will need to stand up thousands of them in a very tight time frame. That screams for a basic power strip – just power in and power out. 

Or does it? 

The no-compromises solution is a power strip that supports remote outlet switching and offers integration through SNMP or via API to most major DCIM and data center management tools.

I’d like to hear from you:

  1. Who do you turn to for advice?
  2. What do you look for in a supplier?
  3. What matters most in the product you choose?

Server Technology has been in the PDU business for over 20 years, and created custom power solutions for most every internet property on the planet in volume and on time, with all the right certifications.  Whether you have a CAD file for a build to print situation or just a sketch on a napkin that you need someone to finish, your Power Strategy Expert from Server Technology can help. Engineers are standing by!

And if you want more of a self-service option, check out the only “Build Your Own PDU” site on the web.

Email me at [email protected] or on twitter @mcram01

Your Hyperscale Data Center Demands Reliability. Why Compromise? Learn More Today

Topics: build your own PDU, Hyperscale

What comes first, the network or the application?

Posted by Marc Cram on May 3, 2017 11:05:22 AM

Businessman hand working with a Cloud Computing diagram on the new computer interface as concept.jpeg

If you believe many of today’s publications, sensor-laden driverless cars look to become a part of everyday life over the next decade. The processing power needed to handle the flood of data for driving, along with vehicle to vehicle, vehicle to highway, and vehicle to dispatch/management communications is likely to be huge. Edge computing, putting compute infrastructure close to the point of use (beside or over the highway, for example) will likely be called for along with deploying 5G wireless communications for transporting the data.

Here are a few questions to ponder:

  • Will the car companies (Ford, GM, Toyota, Hyundai, Tesla) build their own wireless networks to move that data?
  • Will the city and state governments who manage the roads and highways move that data?
  • Will the wireless network owners be responsible for transport?
  • Will a cloud provider such as Amazon or Google build the network that enables Uber and Lyft to function?

 

I’d like to hear your inputs on where we are going and how we get there. Email me at [email protected] and @mcram01 on twitter.

Download the White Paper on Edge Computing to Learn More

Smart Cities Need Fog Computing with Intelligent Power

Posted by Marc Cram on Apr 26, 2017 11:02:06 AM

Businessperson holding media image of city in palm. New technologies-1.jpeg

Mainframes and Moore’s law led to personal computers.  Client-server applications became possible with the first local area networks. Cellular radio systems and Wi-Fi, along with Moore’s law (again) combined with improved battery technology have made laptops, tablets, cell phones, and augmented reality headsets key drivers of internet activity today. Tomorrow’s applications will be more widespread, and possibly less visible. Think smart cities, where the lamp posts and the sidewalks work together to guide you to your destination so you don’t have to watch your progress on a map application on your phone. The solar powered talking trash bin on the corner can call a driverless Lyft for you. Need to make a phone call? Put your hand on the glass of the bus stop shelter and you can have a video call for a few micro-cents.

These distributed “fog” applications need local compute support, and that requires reliable electrical sources to power them along with remote management tools to monitor and report issues to the appropriate companies and agencies.

Are you tasked with enabling these projects?

Would you use this infrastructure if it was available?

I’d like your vision and feedback on this. Drop me a line at [email protected] or @mcram01 on twitter.

Learn More About What We Can Do For Your Edge Computing Application

Topics: intelligent PDUs, smart, fog computing

Three Questions for Your Edge Application

Posted by Marc Cram on Apr 19, 2017 10:44:22 AM

cloud-is-up-city-MOD.png

Your centralized hyperscale data center is up and running in a stable fashion. Now the software team has come up with applications that are so bandwidth intensive that you are going to have to do some extensive pre-processing in every major locale to reduce network traffic and latency times. Sounds like some form of edge computing is needed, whether that is edge, mobile edge or even fog computing.  And wherever distributed/edge computing is called for, intelligent remote power management is a requisite. 

Here are three questions that should accompany every edge computing deployment:

  • How are you going to avoid a truck roll when hardware is locked up?
  • How are you going to know if you are consuming more power in one location than is typical?
  • How are you going to know if the site is having a thermal issue?

Your Power Strategy Experts are standing by to help you answer these questions and more. 

But I’d like to hear from you – what do consider to be the most important factors when designing an edge computing application?

Email me at [email protected], or @mcram01 on twitter.

Learn More About What We Can Do For Your Edge Computing Application

 

Hyperscale Demands...

Posted by Marc Cram on Apr 7, 2017 11:29:07 AM

Utility power should just be there. Always on. Never failing. Today’s hyperscale data center designs frequently count on the electric utility to supply them with a stable source of clean renewable energy. Alternatively, some use locally generated power with the utility as a backup.  Combining robust software stacks that incorporate either virtual machine or container technologies in a rack with in-rack UPS solutions ensures a high degree of uptime and failover capability. Your hyperscale data center can’t tolerate a PDU as the weakest link in the power distribution chain. You need the best PDU solution in the business. You need Server Technology.

What are the most important things you look for in choosing a power solution?

  1. Price?
  2. Availability?
  3. Technology?
  4. Reliability?
  5. Support?

I’d like to hear from you. Email me at [email protected] or on twitter @mcram01

Your Hyperscale Data Center Demands Reliability. Why Compromise? Learn More Today

Topics: Hyperscale

Datacenter Models: Private Cloud

Posted by Marc Cram on Sep 16, 2016 9:43:45 AM

private_cloud-01-2.jpg

Private cloud is a term that is easily misunderstood.  "Private cloud computing is defined by privacy, not location, ownership or management responsibility," Gartner’s Tom Bittman says. A private cloud is dedicated to a single customer, and may reside in customer owned premises, or in the cloud provider’s premises. Operating your IT in a private cloud offers many of the same benefits of public cloud, such as scalability and energy efficiency.

For those having applications that require compliance with HIPAA or PCI DSS, operating a private cloud infrastructure can make a lot of sense.

See the Server Technology Industry Brief “What Type of Datacenter User Are You?” for a further discussion on colocation, public and private clouds.

Datacenter Model Industry Brief

Topics: private cloud, data center options, data center model

Datacenter Models: Public Cloud

Posted by Marc Cram on Sep 14, 2016 3:34:38 PM

private-cloud-01-01.jpg

Most everyone in the IT field today is aware of the availability of public cloud infrastructure from Amazon, Microsoft, Google, IBM/Softlayer, and others. Public cloud services are the most rapidly growing segment of IT according to Gartner and other industry analysts. The benefits of adopting public cloud are numerous, such as scalability, flexibility, and the ability to operate without capital expenditures. In many cases, relying on public cloud can be greener than a custom built datacenter.

For those with growing IT needs and not having room for expansion, migrating applications to a public cloud provider can be a quick, easy, and inexpensive way to deliver the functionality needed without spending a lot of cash up front on hardware.

See the Server Technology Industry Brief “What Type of Datacenter User Are You?” for a further discussion on colocation, public and private clouds.

Datacenter Model Industry Brief

Topics: data center models, public cloud, data center model

What Type of Datacenter User Are You?

Posted by Marc Cram on Sep 12, 2016 10:32:43 AM

datacenter_type-01-1.jpg

While both “cloud” and “cloud-first” are the new go to IT solutions for many companies, there remain a large number of situations where a complete outsourcing of your hardware infrastructure is not practical. In that circumstance, colocation of your IT should make the short list for consideration, whether it is driven by the needs for expansion, proximity, or interconnect. Colocation offers the advantages of highly efficient buildings, support for multiple locations, and access to some of the best interconnections available in the industry.

For those that are expanding beyond the use of a few hundred virtual machines running in the cloud, growing into one or more colocation facilities can bring a quick return on investment without requiring the capital to build your own green field datacenter. Colocation can also serve as a bridge point between public and private cloud deployments.

See the Server Technology Industry Brief “What Type of Datacenter User Are You?” for a further discussion on colocation, public and private clouds.

Datacenter Model Industry Brief

Topics: cloud data center, colocation, cloud first, datacenter model

Got Convergence? Get a Converged PDU

Posted by Marc Cram on Feb 16, 2015 9:41:02 AM

datacenter1

Today, Intel and others are exploring the ramifications of a “disaggregated” server. Putting the RAM on a shelf. The storage on a shelf. The compute on a shelf. Then interconnecting them with a multi-way optical switch that provides connectivity between the shelves as well as off rack communications. The rack becomes a computer with all of the commodity parts providing low cost redundancy and hot-swap capability.

What are the main components of a desktop PC? The hardware consists of a power supply, a motherboard with a CPU, a permanent memory drive (hard drive, optical drive, etc.) to store the operating system, main memory (usually DRAM), a graphics controller that is often integrated onto the motherboard, a network interface, a sound processor/amplifier, and a chassis.

Now take a look at a “(hyper) converged server.”  Let’s see. There is a server-oriented CPU, some permanent storage (hard drive, SSD), a network interface, a power supply, some RAM, a graphics controller, and a chassis. Looks pretty similar overall to the desktop of yore! Even the form factor is similar. It seems that we have come full circle from where we were in the 1990s. The only significant differences from a desktop PC are the scale (more memory, faster drives) and the software stack running on the platform that gives the “hyperconverged server” its native support for virtualization and containers. Presumably then, “hyperconverged” is a comparison to the “disaggregated” server. What “hyperconverged” really has going for it is the ability to put additional units into a rack or datacenter that can be discovered and integrated to operate together in seamless expandability. We definitely couldn’t do that in the 1990s. Every system was a standalone piece of discrete hardware (client-server architecture for those with long memories).

A rack of (hyper) converged appliances are going to draw a fair amount of power, and require a significant number of outlets. A typical 2U appliance will have dual power supplies, each of which requires a C13 outlet. So for a 42U rack, that is 21 appliances having 42 power supplies. In a redundant power configuration, that puts two PDUS of at least 21 outlets each. With a typical 1200W maximum per power supply, that is a potential load of 25.2 KW on either strip. For most datacenters, that means going either with 208V 3-phase 60A PDU (and running the risk of tripping the branch circuit protection), or moving up to 415V circuits of 32A or 60A, requiring some very high capability PDUS.

Someone implementing such a rack of converged products is not going to have any tolerance for downtime. Too many people and processes are all relying on the performance of the hardware in that rack. This means the PDUs powering this gear must be rock solid and capable of communicating whether running on mains or backup power. Check out the new Pro2 architecture from Server Technology. It too is converged, and it delivers!

www.servertech.com

Topics: data center best practices, convergence