Streaming on Disney+ now is Marvel Studios’ What If…? animated TV series. In the first episode, Peggy Carter, instead of Steve Rogers, took the super soldier serum and became the first Avenger. The TV series explores alternatives and possibilities of what we may have considered as precept and the order of things.
As storage practitioners, we are often faced with certain “dogmatic” arguments which were often a mix of measured actuality and marketing magic – aka FUD (fear, uncertainty, doubt). Time and again, we are thrown a curve ball, like “Oh, your competitor can do this. Can you?” Suddenly you are feeling pinned to a corner, and the pressure to defend your turf rises. You fumbled; You have no answer; Game over!
I experienced these hearty objections many times over. The best experience was one particular meeting I had during my early days with NetApp® in 2000. I was only 1-2 months with the company, still wet between the ears with the technology. I was pitching the SnapMirror® to Ericsson Malaysia when the Scandinavian manager said, “I think you are lying!“. I was lost without a response. I fumbled spectacularly although I couldn’t remember if we won or lost that opportunity.
Here are a few I often encountered. Let’s play the game of What If …?
What If …?
Scale-out vs Scale-Up
Perception: Scale-out storage architectures performs better than Scale-up architectures
What If #1: The Scale-out storage performance does not grow linearly for sustained throughput and latency builds, even though the storage capacity growth can be addressed with the scale-out?
What If #2: The Scale-Up storage is able to deliver a more sustained performance for the demanding workloads?
Things to look for: Workload characteristics – random/sequential, IOPS/throughput/latency, block sizes, workflow.
Active-Active HA vs Active-Passive HA
Perception: Active-Active HA pairs perform better than Active-Passive HA pairs in a storage array
What If #1: After a controller has failed in an Active-Active HA, the surviving controller has to balance both controllers’ workload, thereby slowing the performance prowess the HA pair was able to deliver during normal operations?
What If#2: In an Active-Passive HA pair, the surviving controller (previously passive) is able to handle 100% of the performance prowess of the entire storage array, and thus during failover, the operational demands were served without much degradation in performance?
Things to look for: Continuous performance requirements after a failover over a period of time. Impact to the operations and business.
Sync Replication vs Async Replication
Perception: Sync Replication is better protection than Async Replication
What If #1 : Sync Replication is spreading data integrity issues (including ransomware infected files) almost immediately from the production (source) storage volumes to the secondary (target) storage volumes? You get issues on both sides instantly and recovery is affected?
What If #2: The continuous synchronous replication increased the round trip service time in a lesser bandwidth network session, and the database application times out?
What If #2: The time gap depending on the RPO (recovery point objective) is a better measure to dictate if Async Replication might be a more fit into the operations of the business?
Things to look for: RPO and RTO. Network delivery viability.
FUDs of honourable mentions
Here are others I encountered
- Position (not) in the Gartner Magic Quadrant
- Fast IOPS with 10K/15K RPM drives versus Solid State Drives based on price
- Enterprise SATA versus SAS
- Software defined storage vs storage appliance
- Et al which I may have encountered but can’t think of them right now
The safe bet
“Nobody got fired for buying IBM®“. It was a well known phrase in the IT industry back in the 80s and 90s. It is of course no longer true, but from the psychology of the end user, it was a safe bet. It was that “Get out of Jail for Free” card and thus they were (superficially) protected from their fumbles and missteps of their decisions.
Similarly we see end users practice these objections often, partly to be safe and partly to mask their unpracticed inexperience, among other things. Storage practitioners also do the same too, including myself sometimes, placing certain conversations in the form of regurgitation of schools of thoughts that we were coached to pitch. Drinking your own Kool-Aid, I suppose.
But when we look beyond these walls of idiomatic conversations in storage technology, we must look for the “safe bet” buttons of the end users. What makes them feel safe? Why do they say the things they say? How do we earn their trust?
More than meets the eye – charming or offensive
If you have noticed, I did not respond to the Perception with the word Reality. I said What If.
There are many, many variables to these conversations, and a reality to someone is a mirage to another. So, what I am offering here are not solutions and panaceas, but a view of the different schools of thoughts. It is important that storage practitioners and professionals should not be too engrossed and get sucked into these objections-responses dichotomy of which is better. This is not a playground dare of “My whatever is bigger than your whatever“.
How we respond to these “confrontations” is up to us. We can be charming or we can be offensive.
Completing the circle
What is important to take a step back and listen more intimately to the end user’s business and operational challenges. Understand beyond the speeds and feeds, features and functions and look deeper into into the people, and the processes, and leave technology until it mattered further into the conversation.
Dealing with FUDs is energy draining. And the win rate against an incumbent or with an end user who already has a strong perception of a certain storage school of thought is likely to be less encouraging. So turn the conversation around, and look into how well you as a storage practitioner understand their industry, the market they are in, the challenges an issues they face in their short-term and long-term business plans and then bring in your side of the alternative technology to them.
Our value should be beyond the shiny stars, and into the other less shiny things in the end user’s universe. Our value is to solve their operations and business challenges, not to be too caught up in the technology banter and tai-chi.
What if .. I was philosophical about this whole thing. I would probably say,
As within, so without – Hermes Trismegistus
Pingback: Random Short Take #60 | PenguinPunk.net