webMethods Optimize

Issue 2, 2015

Download pdf
 

Three uses that surprised me

Learn how webMethods Optimize is used by customers for more than its original purpose.

A revelation

As product manager for webMethods Optimize, I saw some interesting uses of the product over the years. I have seen it used for everything from tracking server room temperature and saltwater aquarium salinity to just about every conceivable use case in between. Recently I moved into a product marketing role for webMethods Operational Intelligence and have been meeting with a lot of interesting customers with interesting stories. This is the genesis of this story: I have started to see some use patterns of Optimize that are becoming more and more common. The interesting thing to me is that these use cases build upon core Optimize features in ways that I find surprising. 


Original intent

First a little background. If you are a long-time webMethods user, you’ve heard about Optimize for many years. webMethods acquired the Optimize technology long before Software AG acquired webMethods. It was applied to: Optimize for Process, Optimize for Infrastructure, Optimize for SAP®, Optimize for Mainframe and Optimize for B2B. Each of these offerings was customized for the specified task and helped greatly reduce the cost and time required to implement monitoring and alerting solutions. All of these solutions build upon the following core Optimize functions:

  • Event and analytics processor: We call this the Analytic Engine; it’s a high-throughput, low-latency engine that filters incoming event data into Key Performance Indicators (KPIs) and looks for behavioral anomalies.
     
  • Graphical rule processor: This determines the appropriate response to conditions that have been detected.  
     
  • Interactive user dashboards: Optimize ships with near real–time, continuously refreshing dashboards that are customize for their specific use case.  It should be noted that the Optimize engine is easily accessible by other dashboarding tools as well.
     
  • Response manager: Optimize has a very robust response manager that allows it to start orchestrated processes, call Web services or send Event Driven Architecture (EDA) events. It also ships with a complete task list management system for coordinating resolutions that include human actions.
     
  • Alert manager: This facilitates sending alerts, such as email and text messages.


Customer-innovated use cases

As the Internet of Things has started to mature, there is a LOT more data being generated by fringe devices. Companies are trying to find innovative ways to deal with this onslaught of data.

Here are the top three surprising use cases—ways customers are using webMethods Optimize to deal with the new challenges of the Internet of Things:  

  1. Data store
    Even though Optimize is a kick-butt Business Activity Management (BAM) system, some customers have thrown this capability to the wind and instead use Optimize as a very fast, easy–to-configure data store. Optimize can easily capture just about any data metric whether it is from a process, integration server or universal messaging. The data streams in Optimize sort the data, baseline it and then store it in a database table.

    In this scenario, the customer does not typically care about the sophisticated baselining and alerting capabilities of Optimize. Instead, they pull tables from the database to run the data against their own business intelligence tools. While I might think that this is like using a car to drive 10 feet, customers argue that Optimize does such a good job as a data storage product that it has become fundamental to their business.
     
  2. General purpose alerting engine   
    Some customers use Optimize strictly as an alerting engine and forego the interesting data that can be derived from the stored metrics. While there are many potential solutions for alerting in an organization, I have seen more and more customers use Optimize as a very convenient business-user focused tool that can alert on virtually any type of streaming data.

    Please don’t confuse this use with Apama’s world-class Complex Event Processing (CEP) engine. There is nothing really complex with the alerts that many of these customers are using Optimize for. Yes, Optimize specializes in alerting on statistical and behavioral anomalies within streams of data, but I see it used for simple event data such as: “Send me an email whenever an event is detected on the event bus that has a value of X for this customer.” Complex? No. Valuable for a business user? Absolutely. Again, I might consider this a misuse of the product, but if customers have discovered a valuable use for this product, then who am I to judge?
     
  3. Separating the wheat from the chaff (not literally)
    If you have followed my writings in TECHniques, then you know that I absolutely love EDA and think that it will revolutionize the way we integrate systems in the future. In many ways, the Internet of Things has helped drive the adoption of EDA as customers try to deal with the ever-increasing volume of edge data being delivered to the enterprise. The problem some customers experience is there is so much data coming in that they don’t know what to focus on. Optimize to the rescue.

    Optimize is specifically tuned to baseline the performance of any metric and tell you when something is not normal. This feature has been used for a long time in Optimize for Infrastructure for such things as memory utilization or queue depth. Only recently have I seen this capability applied to things like hardware environmental metrics (such as temperature and humidity), manufacturing lines (for volume and velocity, for example) and human behavioral monitoring (such as crowd analysis and path utilization).

These use cases all have something in common: The customer does not care about the actual number, but rather the variance from normal. Do I care that the temperature in a smoke stack is 127 degrees? Maybe, especially if the normal temperature for a Tuesday at 2 p.m. is 243 degrees. In these types of cases, Optimize serves as a context engine that allows the user to send alerts on understood behavioral variances. 

While we may have developed webMethods Optimize as a BAM tool, I now understand that it has applicability in many different use cases and complements the Apama CEP engine. Optimize delivers an interesting perspective to data metrics that is unique to the product.

Is your company using Optimize in an innovative way? I would love to hear about it and maybe develop a blog post about it. Comment on this blog post or email me with your story.