Future of EAI

The business point of view in enterprise application integration looks bright as ever. In companies old school approach “let’s consolidate all to one huge ERP” have caused too much collateral damage and costs by offering almost good solutions to business needs. Best of breed approach, where business choose best system suitable for their needs, is gaining ground in majority of companies. Best of breed thinking, and especially buying, leads to more heterogeneous system infrastructure and therefore increases markets of EAI. 

Another big factor in future will be the fact that cloud hype is slowly starting to actualize in companies. For example, cloud-only CRM Salesforce is leading the way and today every CRM has cloud service option. ERPs are doing the same. The slowness of cloud-adoption is not slow because of the markets are lacking services. Quite naturally it is the large enterprises that just move slowly to cloud. In addition, not everything cannot be transferred to cloud due to the local legislation etc. so companies are heading towards on-premises - cloud -hybrid solutions where weight is generally first at on-premises and later on weight transfers to cloud. Cloud adoption again increases the need for integration from on-premises to cloud systems and there will be more and more cloud-to-cloud integrations between internal systems and between business partners.

Cloud has also a big risk: cloud spaghetti. As more and more cloud services offer direct point-to-point integration channels to other major cloud-based enterprise systems, it is tempting to just click’n’connect all these. No problems arise when there is only few systems point-to-point integrated. But think forward: The number of connections needed to have fully connected point-to-point environment, with n systems, is given by

\frac{n(n-1)}{2}

Ten cloud applications to be totally integrated point-to-point,it requires

\frac{10\times9}{2}

spaghetticloud_2

connections totaling 45. This is of course super-pessimistic worst-case scenario, but you’ll get the point – try to change one system in a spaghetti and changes will propagate everywhere. It is obvious that before we reach 10, we’ll bump into classic EAI problems like differences in interpretation of standards, need for enrichment of data, customization etc. From process integration view this is like going back to stone-age – no business process monitoring on top of integration platform there.

Estimated need for cloud integration has led to recent boost of cloud integration providers. As large enterprises can buy suitable best-of-breed cloud services for certain needs, it is not probable that they would trust the business critical integrations to such providers. In the other hand, small cloud integrators can offer price-level and services what small and midsize companies can afford and need. This actually opens the capabilities of integration to new market segment that earlier could not afford integration platform and implementation investment.

Overall – integration market is growing due to the Best-of-breed thinking and current cloud adoption. B-of-b and cloud adoption go nicely together as it is easy to deliver b-of-b solutions via cloud. Cloud also leverages the integration markets to smaller companies than before.

2 thoughts on “Future of EAI

  1. Pingback: EAI – a commodity | Integration War Stories

  2. Pingback: 2014: Annoying cloud and mobile backends | Integration War Stories

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s