Archive for the ‘engineering’ Category

assessment and capacity analysis and planning for virtualization initiatives

Thursday, August 25th, 2011

Q:

What would need to go into an assessment tool for a virtualization initiative?

A:

Typical factors may include:

  • current CPU load per server
  • what’s running on each server
  • current hardware of each server
  • expected percentage increase in usage
  • OS usage – homogenous or heterogeneous
  • new hardware or re-use current hardware
  • storage needs
  • vendor for virtualization (VMware, Microsoft, Xen)

And don’t forget the all-important:

  • BUDGET

My experience is all related around VMware, but what I’ve seen and used in the past is the following:

  • look at all CPU utilizations currently
  • add those average and peak percentages in two separate columns
  • plan for ~10% overhead from your hypervisor of choice
  • for every 40% of ‘average’ or 80% of ‘peak’, use one server of the type you now consider “high-end” (ie, if you have a total of 687% of ‘peak’, you need 9 physical servers running your hypervisor of choice)

Other thoughts:

  • I like to plan for 1 full spare physical server per ~6, so that I can utilize Vmware’s Vmotion for migrating servers around
  • plan for buying/utilizing SAN storage of some form so your VMs can be moved to different physical servers easily

I originally answered this topic ~2 years ago on serverfault.com.

chimneys and fireplaces

Sunday, May 29th, 2011

About 14 years ago, good friends of the family bought an old farm house with some outbuildings in rural Schoharie County in NY (outside Middleburgh, in a locale called “Hunter’s Land”).

One of the first things they wanted to do with the house was to re-engineer the heating, which included properly insulating the house, and then adding a back-to-back fireplace and stove nook in the living room, backed to the kitchen (the nook was designed to look like it had once been a fireplace). Jeff holds a PhD in Chemical Engineering, and works full-time with fluid dynamics for a government contractor. So when he deep-dove into fireplace fundamentals, fluid flow mechanics, and heat transfer, all of us who knew him well knew he’d come out with the best available design for heating his new home.

The design Jeff eventually settled on was a modernized Rumford fireplace (Count Rumford was quite the experimenter). One of the major differences between a Rumford design and that of a “traditional” fireplace is the smoke chamber is reversed – ie, the slope goes from front-to-back, and not back-to-front (see diagram, provided by McNear).

fireplace diagrams

comparison diagram of Rumford and traditional fireplaces

The primary advantages to this reversal are: reduced/eliminated smoke discharge into the room (especially at low-flow heating circumstances), and simplified construction (no corbeling of the back wall). Other improvements in the design are broader radiatory angles for heat dispersion, and reduced fuel consumption.

Fortunately for Jeff and his wife, his brother-in-law was a mason, so costs could be reduced for installing the chimney and fireplaces to just that of labor and materials: still an expense, but lower than hiring it done outside the family.

Also fortunately for them, they had a ready supply of friends to help tear-out the non-load-bearing wall separating the kitchen and living room, cut a hole in the floor, pour a foundation for the structure, and build-out the subhearth.

During that period, I was going weekly to their house where Jeff and I would spend most of the evening geeking-out: I was in high school, and Jeff was teaching me how to program in C++ (well, ok: he taught me how to program in C using C++ keywords, and then I taught him object-orientation). His day job involved utilizing both commercial and in-house finite element analysis tools (utilizing a [then relatively] new equation solver: GMRES), and he used the time with me to try-out ideas he might want to incorporate at work – just in a new/easier-to-use language than that which he employed from 9-5 (his work centered around Fortran). (As a sidebar, we both learned an enormous amount during those days – good times [even ultimately leading to the Story of Mr G {the second version of the story}]). Back to the story.

Installing the fireplace was a fairly straight-forward process: once the wall and floor were down and cut (and the support beams for the house re-supported on the fringe of the subhearth and the middles cut out, the fireplace was laid-up “normally”. Thanks to pre-cast pieces from Superior Clay, installing the throat and smoke chamber was a cinch. Bringing the flue and chimney up through the roof was likewise a pretty quick process (the portion of the house the chimney went did not have a second storey), culminating in the first celebratory fire about a year after the whole process started (Jeff worked on it part-time during evenings and weekends, getting help as he could).

During that time, I learned a buttload about masonry, the densities of limestone and cast iron (the mantle and stove), re-engineering existing homes, temperature profiles of a fire, and fluid dynamics in action. Due my current occupation at the time, I had access to a host of thermocouple devices, some of which I checked-out and brought to Jeff’s house to profile the fireplace for where it was hottest, so that fires could be built (and a reflector plate installed) in the most efficient manner.

After a few years in Hunter’s Land, Jeff & co moved to a new domicile closer to his work (cutting his ~60 minutes each way commute by 80%). They again decided to add a Rumford fireplace to their home, but hired the entire project out for time purposes.

Since that time, I have been a proponent of fireplace heating, and of some level of self-reliance on fuel (and food) supplies (when possible).

Who knew a year-long hobby project would end up having such a long-term effect?

Given the chance, I’d LOVE to implement such a fireplace in my own home when we finally buy one.