Infinite I/O

4 Must Do's Before You Evaluate New Tech

Posted by Matt Brender on Dec 16, 2014
Find me on:

It's that time of year where new toys may be on your mind. The holidays and end of year budgets overlap into a time of professional joy. Sometimes red tape drops and POs go through faster than ever. But let's not forget some of the basics in the rush of excitement! 

Nearly all Infinio customers evaluate our software before choosing it. Here are some of the things we've seen this year that have lead to our most productive evaluations.



Step 1: Have the basics looked over

Nothing brings the excitement of getting new tech to a halt like uncovering an old issue you didn't know you had. It's worth taking half a day and cleaning up after the long year of action items. When you think about your virtualized environment, don't forget these basics:

  • Is DNS resolution error free?
  • Is vCenter error free and healthy?
  • Are my storage, vMotion and management networks properly isolated?

If any of these answers are "no," then you have plenty of work to clean up before you want to bring in another tool. If you're curious about the importance of the last question, reread this post.

Step 2: Define meaningful metrics

Few POs go through because someone says "I want this software." Most bosses and colleagues like to understand why one toy is more useful than another. To have a meaningful comparison, you need to have a measurement that's comparable across all products

Will this purchase go to the least expensive option?

The longest lasting?

Does it need to reach a certain level of scale and stay below a certain latency threshold?

Take a few minutes and write down the leading decision making criteria. It's time well taken and will speed up the rest of the process.

Step 3: State your business objective

Excellent - we have metrics that everyone cares about. But does those objectives map to the business requirements? This step takes the metrics defined above and puts them into sentences. For instance, if money is tight but storage performance is an issue, try:
  • I will choose the least expensive solution that enables us to push ____ IOPS from ____ VMs and maintain no higher than ____ latency.
So as a full sentence:
  • I will choose the least expensive solution that enables us to push 10,000 8KB IOPS from 20 VMs and maintain no higher than 15ms latency.

We've now gone from a "nice to have" conversation to an "if this, then that" decision making process. And that's ideal. 

Step 4: Remember the big picture

As much as exciting nerd knobs and shiny new things are a fun part of the job, everyone will appreciate you more if you remember what constraints you have to work within. For instance, remembering that: 

  • Downtime is costly
  • Data migration is time consuming
  • Operational change is cumbersome

Taking a little bit of time before you pursue a big new project can work wonders for your goal of getting what you want and for the team.

Start Your Trial of Infinio

No matter how many new data center toys you have headed your direction this holiday season, enjoy it and Happy New Year!

Topics: Talking Tech