Area Title System (DNS) is rightly celebrated because the protocol that makes the web attainable. DNS is named the phonebook of the web as a result of it interprets domains to IP addresses so browsers can load web assets. It normally works sufficiently sufficient that the majority community groups barely pause to consider it.
But DNS additionally has the doubtful distinction of being the protocol that breaks the web or slows it down. “It’s at all times DNS” isn’t only a meme or a joking clarification for downtime that seems on Reddit threads. It’s an expression of actuality: the blame for poor software efficiency is commonly traced again to points with underlying DNS methods.
The problem of DNS efficiency measurement
Most community groups perceive that the way in which they deploy authoritative DNS issues, however solely in a conceptual means. They know that managed DNS providers supply quicker response instances and enhance software efficiency, however there’s no onerous information to again that up. Surprisingly, most community groups make choices a couple of important system that powers their whole enterprise utilizing principally guesswork and gut-level instinct.
Some community groups use websites like DNSPerf to benchmark the efficiency of managed authoritative DNS options, however they don’t actually delve deeply into what the numbers on that web site imply or how they’re calculated. Additionally they can’t use the info on that web site to check managed DNS options and self-hosted DNS.
This lack of dependable, comparative information on managed DNS options has been a recognized hole for a very long time. Right here at IBM®, we’re consistently requested about how our IBM NS1 Join® answer for managed DNS is totally different from all the opposite choices within the authoritative DNS market. Community groups wish to know whether it is actually going to make a distinction within the efficiency of their functions or web sites.
A joint IBM-Catchpoint examine
Given all of the questions we get from clients and prospects on this problem, we’ve determined that it was time to fill this hole ourselves. We reached out to Catchpoint, our long-time collaborator and the consultants in community efficiency measurement, to see what we might study.
That outreach led each IBM and Catchpoint down an interesting street. We discovered that an excessive amount of of the info on authoritative DNS efficiency floating across the web lacked statistical rigor. It wasn’t consultant of various geographies or market verticals. The info had too many technical asterisks by it. It didn’t present an apples-to-apples comparability as a result of there are simply so some ways to configure DNS.
After working by varied analysis choices and operational pathways, we determined to go huge and design our personal examine of DNS efficiency from scratch. IBM constructed out an inventory that represents over 2,500 of the biggest corporations on the earth. The pattern represents DNS utilization throughout geographies, market verticals and authoritative DNS deployment sorts. Catchpoint created a efficiency take a look at, which stripped out all the varied DNS accelerators and intermediate layers. This resulted in a typical baseline measurement that may be relevant to any enterprise.
We ran checks within the busiest season of the 12 months for many companies, November and December, to seize the interval the place most enterprises require the very best DNS efficiency. The info was captured over the course of a number of weeks, so we might see the ebb and move of web visitors over time relatively than counting on much less consultant spot checks.
What we discovered
When the outcomes got here again, they demonstrated very clear outcomes. Like most community groups, we had assumptions concerning the relative efficiency of managed DNS options versus self-hosted methods. These assumptions weren’t solely verified, however the hole turned out to be a lot bigger than we realized. Spoiler alert: managed DNS options are quicker—quite a bit quicker.
We have been additionally naturally curious concerning the efficiency of the IBM NS1 Join answer in opposition to different authoritative DNS choices. We bought some fascinating information on this space as properly, a few of which stunned us.
Past the comparative efficiency of DNS options, we additionally discovered quite a bit about variations in web efficiency between totally different geographies and market verticals. It seems that the place you’re accessing the web can have simply as a lot to do with efficiency as which firm’s functions or web sites you’re accessing.
For those who’re at the moment internet hosting your personal authoritative DNS and are interested by shifting to a managed DNS answer, or if you’re making an attempt to check distributors available on the market, you’re going to wish to see this information. This examine will even be of curiosity in the event you’re simply fascinated with benchmarking the efficiency of your DNS in opposition to a dependable customary.
Learn our white paper
Was this text useful?
SureNo
