Cutting Costs in the World of Boomi

Hello Boomi fans!

Economic activity is slowing down.

Many companies are downsizing, even the MAG 7 tech giants.

At the same time vendors are raising their prices.

72% of global companies exceeded their cloud budgets.

As a result, 83% of enterprises plan to move workloads back to Private DC.

Across the board, departments are asked to cut costs.

The integration department is no exception.

On the other hand, the need for integration is exploding.

So how can we do more with less?

Cutting costs - Case in Point:

A Boomi customer contacted us because they had a hard time keeping their integration services stable at 6.5 Million process executions / month.

One of the key concerns was the frequent execution spikes of 1250 executions / minute.

The Customer was encountering monthly outages. The setup and configurations were not coping with the load.

Our customer was bracing for an upgrade from 1 Atom to a 3 Node Molecule (or more).

Let's estimate the extra costs in USD:

+ 3 X the cost for the servers = ~ 60K / year

+ cost for a high end NFS solution & backup = ~ 150K / year

+ 2 X the admin and maintenance cost = ~ 40K / year

+ Load balancer

+ API Management Solution = ~ 30K / year

+ cost for the installation and migration project = 100 K

Projected Total Costs = 940K over a period of 3 years.

But we know that migrating instability from one setup to another doesn’t make the end solution more stable.

I proposed to secure the integration landscape first.

This is how we helped our customer stabilize their environment and avoid the migration costs entirely:

We introduced throttling in multiple layers: queues, web server and global maximum amounts: we “rev” Boomi all the way to “red”, but we do not go on “red” to avoid outages.

We identified and removed performance hot spots like:

• 1 process that was consuming 56% of the storage available for Boomi.
• 1 source system was not respecting the SLAs and producing severe request spikes that were endangering the stability of the entire platform.
• 1 target system that was periodically slowing down consumption due to internal batch processes. This was causing message pileup in Boomi.
• We did performance tuning for the runtime configurations to align the runtime with the work load patterns.
• and many, many more adjustments.

The Projected Total Costs was estimated at 940K over a period of 3 years.

The Actual Cost for remediation (that had to be done anyway) was 100K.

Costs avoided = 840K. So the ROI = 840 %

We estimate that the improvements we made will allow the Customer to operate as it is, for the next 3 years, given the current state and the current rate of change.

How did we know what needed to be done?

We used our Monitoring + Analytics Solution to profile the Boomi Runtime and Processes Executions and build the list of necessary improvements, ordered by ROI.

This is how we give you the biggest bang for your buck.

Request a Demo Today !

~~~~~~~~~~~~~~~ Now go #BoomIT ! ~~~~~~~~~~~~~~