Jul 18, 2005This article was originally published by RFID Update.
July 18, 2005—People often scratch their heads when they learn of my two areas of research focus: RFID technology and IT Governance. On the surface, they seem to be at opposite ends of the spectrum and I could probably write a whole separate column about how my background has brought the two together. Fortunately, I will spare you those gory details and focus on how the RFID world has much to learn from IT, specifically the people and technology responsible for IT Systems Management.
Does this sound familiar?
An item is supposed to be moved from location A to location C through location B. The item makes it to location B but sits there, resulting in a missed commitment to location C causing lost revenue or higher costs because you have to send the item again. Am I talking about a Consumer Goods company shipping to Wal-Mart? Actually no, I was referring to an application (location C) that was requesting data from a database (location A) that traveled through a message queue (location B). The parallels between the movement of goods through a physical supply chain and the movement of packets through a network are striking. As a result, the RFID world has a lot to learn from what works and what doesn't in the world of systems management specifically in these three areas:
- Monitoring – RFID deployments are largely still single site pilots. With the maturation of retailer mandates, the number of sites where RFID is deployed will grow dramatically over the next 5 years. It is critical for RFID vendors to learn from the world of distributed systems management and leverage standards and incumbent technologies wherever possible. Companies don't want a separate infrastructure to manage RFID devices.
- Correlation and root cause analysis – Beyond just core monitoring, it is critical for IT systems to look at various data inputs from related incidents and identify the root cause of an outage. IT organizations that excel at correlation and root cause are better at proactively addressing issues and improving IT performance and availability. The same notions of correlation and root cause analysis are being applied in RFID to address significant supply chain issues like out of stock.
- Event management – Once the root cause has been identified, leading edge IT organizations have mature processes and automation to identify the right people to resolve the issues; reducing downtime and improving availability. In the RFID world, people like to talk about sense and respond. The easy part, and where people are focused today, is on using RFID data to sense changes to product flow. The harder part is implementing the processes and technologies to effectively respond to changes in demand. This is at the heart of what AMR Research calls the Demand-Driven Supply Network (DDSN).
Beware the Christmas Tree Effect
A common phrase in the IT management world is called "the Christmas tree effect" which results when all the events throughout the infrastructure show up on an event console (with red, yellow, and green indicators). This information overload essentially paralyzes the IT organization because they don't know which events to address first or whether individual events are causing outright downtime or just performance degradation. Emerging IT best practices, often labeled Business Service Management, adds a business context to events to help filter the alerts and prioritize response. The potential for the Christmas tree effect is very high in RFID as well so it is critical for RFID deployments to apply business context at each part of the distributed infrastructure so that the information delivered to employees is actionable.