Click here to close now.



Welcome!

IBM Cloud Authors: Pat Romanski, Sarah Patrick, Liz McMillan, Carmen Gonzalez, Jnan Dash

Related Topics: Java IoT, Microservices Expo, Adobe Flex, IoT User Interface, Apache

Java IoT: Article

Why Averages Are Inadequate, and Percentiles Are Great

Averages are ineffective because they are too simplistic and one-dimensional

Anyone who ever monitored or analyzed an application uses or has used averages. They are simple to understand and calculate. We tend to ignore just how wrong the picture is that averages paint of the world. To emphasis the point let me give you a real-world example outside of the performance space that I read recently in a newspaper.

The article was explaining that the average salary in a certain region in Europe was 1900 Euro's (to be clear this would be quite good in that region!). However when looking closer they found out that the majority, namely 9 out of 10 people, only earned around 1000 Euros and one would earn 10.000 (I over simplified this of course, but you get the idea). If you do the math you will see that the average of this is indeed 1900, but we can all agree that this does not represent the "average" salary as we would use the word in day to day live. So now let's apply this thinking to application performance.

The Average Response Time
The average response time is by far the most commonly used metric in application performance management. We assume that this represents a "normal" transaction, however this would only be true if the response time is always the same (all transaction run at equal speed) or the response time distribution is roughly bell curved.

A Bell curve represents the "normal" distribution of response times in which the average and the median are the same. It rarely ever occurs in real applications

In a Bell Curve the average (mean) and median are the same. In other words observed performance would represent the majority (half or more than half) of the transactions.

In reality most applications have few very heavy outliers; a statistician would say that the curve has a long tail. A long tail does not imply many slow transactions, but few that are magnitudes slower than the norm.

This is a typical Response Time Distribution with few but heavy outliers - it has a long tail. The average here is dragged to the right by the long tail.

We recognize that the average no longer represents the bulk of the transactions but can be a lot higher than the median.

You can now argue that this is not a problem as long as the average doesn't look better than the median. I would disagree, but let's look at another real-world scenario experienced by many of our customers:

This is another typical Response Time Distribution. Here we have quite a few very fast transactions that drag the average to the left of the actual median

In this case a considerable percentage of transactions are very, very fast (10-20 percent), while the bulk of transactions are several times slower. The median would still tell us the true story, but the average all of a sudden looks a lot faster than most of our transactions actually are. This is very typical in search engines or when caches are involved - some transactions are very fast, but the bulk are normal. Another reason for this scenario are failed transactions, more specifically transactions that failed fast. Many real-world applications have a failure rate of 1-10 percent (due to user errors or validation errors). These failed transactions are often magnitudes faster than the real ones and consequently distorted an average.

Of course performance analysts are not stupid and regularly try to compensate with higher frequency charts (compensating by looking at smaller aggregates visually) and by taking in minimum and maximum observed response times. However we can often only do this if we know the application very well, those unfamiliar with the application might easily misinterpret the charts. Because of the depth and type of knowledge required for this, it's difficult to communicate your analysis to other people - think how many arguments between IT teams have been caused by this. And that's before we even begin to think about communicating with business stakeholders!

A better metric by far are percentiles, because they allow us to understand the distribution. But before we look at percentiles, let's take a look a key feature in every production monitoring solution: Automatic Baselining and Alerting.

Automatic Baselining and Alerting
In real-world environments, performance gets attention when it is poor and has a negative impact on the business and users. But how can we identify performance issues quickly to prevent negative effects? We cannot alert on every slow transaction, since there are always some. In addition, most operations teams have to maintain a large number of applications and are not familiar with all of them, so manually setting thresholds can be inaccurate, quite painful and time-consuming.

The industry has come up with a solution called Automatic Baselining. Baselining calculates out the "normal" performance and only alerts us when an application slows down or produces more errors than usual. Most approaches rely on averages and standard deviations.

Without going into statistical details, this approach again assumes that the response times are distributed over a bell curve:

The Standard Deviation represents 33% of all transactions with the mean as the middle. 2xStandard Deviation represents 66% and thus the majority, everything outside could be considered an outlier. However most real world scenarios are not bell curved...

Typically, transactions that are outside two times standard deviation are treated as slow and captured for analysis. An alert is raised if the average moves significantly. In a bell curve this would account for the slowest 16.5 percent (and you can of course adjust that); however; if the response time distribution does not represent a bell curve, it becomes inaccurate. We either end up with a lot of false positives (transactions that are a lot slower than the average but when looking at the curve lie within the norm) or we miss a lot of problems (false negatives). In addition if the curve is not a bell curve, then the average can differ a lot from the median; applying a standard deviation to such an average can lead to quite a different result than you would expect. To work around this problem these algorithms have many tunable variables and a lot of "hacks" for specific use cases.

Why I Love Percentiles
A percentile tells me which part of the curve I am looking at and how many transactions are represented by that metric. To visualize this look at the following chart:

This chart shows the 50th and 90th percentile along with the average of the same transaction. It shows that the average is influenced far mor heavily by the 90th, thus by outliers and not by the bulk of the transactions

The green line represents the average. As you can see it is very volatile. The other two lines represent the 50th and 90th percentile. As we can see the 50th percentile (or median) is rather stable but has a couple of jumps. These jumps represent real performance degradation for the majority (50%) of the transactions. The 90th percentile (this is the start of the "tail") is a lot more volatile, which means that the outliers slowness depends on data or user behavior. What's important here is that the average is heavily influenced (dragged) by the 90th percentile, the tail, rather than the bulk of the transactions.

If the 50th percentile (median) of a response time is 500ms that means that 50% of my transactions are either as fast or faster than 500ms. If the 90th percentile of the same transaction is at 1000ms it means that 90% are as fast or faster and only 10% are slower. The average in this case could either be lower than 500ms (on a heavy front curve), a lot higher (long tail) or somewhere in between. A percentile gives me a much better sense of my real world performance, because it shows me a slice of my response time curve.

For exactly that reason percentiles are perfect for automatic baselining. If the 50th percentile moves from 500ms to 600ms I know that 50% of my transactions suffered a 20% performance degradation. You need to react to that.

In many cases we see that the 75th or 90th percentile does not change at all in such a scenario. This means the slow transactions didn't get any slower, only the normal ones did. Depending on how long your tail is the average might not have moved at all in such a scenario.!

In other cases we see the 98th percentile degrading from 1s to 1.5 seconds while the 95th is stable at 900ms. This means that your application as a whole is stable, but a few outliers got worse, nothing to worry about immediately. Percentile-based alerts do not suffer from false positives, are a lot less volatile and don't miss any important performance degradations! Consequently a baselining approach that uses percentiles does not require a lot of tuning variables to work effectively.

The screenshot below shows the Median (50th Percentile) for a particular transaction jumping from about 50ms to about 500ms and triggering an alert as it is significantly above the calculated baseline (green line). The chart labeled "Slow Response Time" on the other hand shows the 90th percentile for the same transaction. These "outliers" also show an increase in response time but not significant enough to trigger an alert.

Here we see an automatic baselining dashboard with a violation at the 50th percentile. The violation is quite clear, at the same time the 90th percentile (right upper chart) does not violate. Because the outliers are so much slower than the bulk of the transaction an average would have been influenced by them and would not have have reacted quite as dramatically as the 50th percentile. We might have missed this clear violation!

How Can We Use Percentiles for Tuning?
Percentiles are also great for tuning, and giving your optimizations a particular goal. Let's say that something within my application is too slow in general and I need to make it faster. In this case I want to focus on bringing down the 90th percentile. This would ensure sure that the overall response time of the application goes down. In other cases I have unacceptably long outliers I want to focus on bringing down response time for transactions beyond the 98th or 99th percentile (only outliers). We see a lot of applications that have perfectly acceptable performance for the 90th percentile, with the 98th percentile being magnitudes worse.

In throughput oriented applications on the other hand I would want to make the majority of my transactions very fast, while accepting that an optimization makes a few outliers slower. I might therefore make sure that the 75th percentile goes down while trying to keep the 90th percentile stable or not getting a lot worse.

I could not make the same kind of observations with averages, minimum and maximum, but with percentiles they are very easy indeed.

Conclusion
Averages are ineffective because they are too simplistic and one-dimensional. Percentiles are a really great and easy way of understanding the real performance characteristics of your application. They also provide a great basis for automatic baselining, behavioral learning and optimizing your application with a proper focus. In short, percentiles are great!

More Stories By Michael Kopp

Michael Kopp has over 12 years of experience as an architect and developer in the Enterprise Java space. Before coming to CompuwareAPM dynaTrace he was the Chief Architect at GoldenSource, a major player in the EDM space. In 2009 he joined dynaTrace as a technology strategist in the center of excellence. He specializes application performance management in large scale production environments with special focus on virtualized and cloud environments. His current focus is how to effectively leverage BigData Solutions and how these technologies impact and change the application landscape.

Comments (1) View Comments

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


Most Recent Comments
rtalexander 11/21/12 12:58:00 AM EST

Hey, could you post a reference or two that covers the theory and/or practicalities of the approach you describe?

Thanks!

@ThingsExpo Stories
As enterprises work to take advantage of Big Data technologies, they frequently become distracted by product-level decisions. In most new Big Data builds this approach is completely counter-productive: it presupposes tools that may not be a fit for development teams, forces IT to take on the burden of evaluating and maintaining unfamiliar technology, and represents a major up-front expense. In his session at @BigDataExpo at @ThingsExpo, Andrew Warfield, CTO and Co-Founder of Coho Data, will dis...
SYS-CON Events announced today that Fusion, a leading provider of cloud services, will exhibit at SYS-CON's 18th International Cloud Expo®, which will take place on June 7-9, 2016, at the Javits Center in New York City, NY. Fusion, a leading provider of integrated cloud solutions to small, medium and large businesses, is the industry's single source for the cloud. Fusion's advanced, proprietary cloud service platform enables the integration of leading edge solutions in the cloud, including clou...
With the Apple Watch making its way onto wrists all over the world, it’s only a matter of time before it becomes a staple in the workplace. In fact, Forrester reported that 68 percent of technology and business decision-makers characterize wearables as a top priority for 2015. Recognizing their business value early on, FinancialForce.com was the first to bring ERP to wearables, helping streamline communication across front and back office functions. In his session at @ThingsExpo, Kevin Roberts...
SYS-CON Events announced today that Commvault, a global leader in enterprise data protection and information management, has been named “Bronze Sponsor” of SYS-CON's 18th International Cloud Expo, which will take place on June 7–9, 2016, at the Javits Center in New York City, NY, and the 19th International Cloud Expo, which will take place on November 1–3, 2016, at the Santa Clara Convention Center in Santa Clara, CA. Commvault is a leading provider of data protection and information management...
SYS-CON Events announced today that Alert Logic, Inc., the leading provider of Security-as-a-Service solutions for the cloud, will exhibit at SYS-CON's 18th International Cloud Expo®, which will take place on June 7-9, 2016, at the Javits Center in New York City, NY. Alert Logic, Inc., provides Security-as-a-Service for on-premises, cloud, and hybrid infrastructures, delivering deep security insight and continuous protection for customers at a lower cost than traditional security solutions. Ful...
SYS-CON Events announced today that VAI, a leading ERP software provider, will exhibit at SYS-CON's 18th International Cloud Expo®, which will take place on June 7-9, 2016, at the Javits Center in New York City, NY. VAI (Vormittag Associates, Inc.) is a leading independent mid-market ERP software developer renowned for its flexible solutions and ability to automate critical business functions for the distribution, manufacturing, specialty retail and service sectors. An IBM Premier Business Part...
The cloud promises new levels of agility and cost-savings for Big Data, data warehousing and analytics. But it’s challenging to understand all the options – from IaaS and PaaS to newer services like HaaS (Hadoop as a Service) and BDaaS (Big Data as a Service). In her session at @BigDataExpo at @ThingsExpo, Hannah Smalltree, a director at Cazena, will provide an educational overview of emerging “as-a-service” options for Big Data in the cloud. This is critical background for IT and data profes...
With an estimated 50 billion devices connected to the Internet by 2020, several industries will begin to expand their capabilities for retaining end point data at the edge to better utilize the range of data types and sheer volume of M2M data generated by the Internet of Things. In his session at @ThingsExpo, Don DeLoach, CEO and President of Infobright, will discuss the infrastructures businesses will need to implement to handle this explosion of data by providing specific use cases for filte...
Fortunately, meaningful and tangible business cases for IoT are plentiful in a broad array of industries and vertical markets. These range from simple warranty cost reduction for capital intensive assets, to minimizing downtime for vital business tools, to creating feedback loops improving product design, to improving and enhancing enterprise customer experiences. All of these business cases, which will be briefly explored in this session, hinge on cost effectively extracting relevant data from ...
SYS-CON Events announced today that Interoute, owner-operator of one of Europe's largest networks and a global cloud services platform, has been named “Bronze Sponsor” of SYS-CON's 18th Cloud Expo, which will take place on June 7-9, 2015 at the Javits Center in New York, New York. Interoute is the owner-operator of one of Europe's largest networks and a global cloud services platform which encompasses 12 data centers, 14 virtual data centers and 31 colocation centers, with connections to 195 ad...
Most people haven’t heard the word, “gamification,” even though they probably, and perhaps unwittingly, participate in it every day. Gamification is “the process of adding games or game-like elements to something (as a task) so as to encourage participation.” Further, gamification is about bringing game mechanics – rules, constructs, processes, and methods – into the real world in an effort to engage people. In his session at @ThingsExpo, Robert Endo, owner and engagement manager of Intrepid D...
Eighty percent of a data scientist’s time is spent gathering and cleaning up data, and 80% of all data is unstructured and almost never analyzed. Cognitive computing, in combination with Big Data, is changing the equation by creating data reservoirs and using natural language processing to enable analysis of unstructured data sources. This is impacting every aspect of the analytics profession from how data is mined (and by whom) to how it is delivered. This is not some futuristic vision: it's ha...
WebRTC has had a real tough three or four years, and so have those working with it. Only a few short years ago, the development world were excited about WebRTC and proclaiming how awesome it was. You might have played with the technology a couple of years ago, only to find the extra infrastructure requirements were painful to implement and poorly documented. This probably left a bitter taste in your mouth, especially when things went wrong.
Learn how IoT, cloud, social networks and last but not least, humans, can be integrated into a seamless integration of cooperative organisms both cybernetic and biological. This has been enabled by recent advances in IoT device capabilities, messaging frameworks, presence and collaboration services, where devices can share information and make independent and human assisted decisions based upon social status from other entities. In his session at @ThingsExpo, Michael Heydt, founder of Seamless...
The IoT's basic concept of collecting data from as many sources possible to drive better decision making, create process innovation and realize additional revenue has been in use at large enterprises with deep pockets for decades. So what has changed? In his session at @ThingsExpo, Prasanna Sivaramakrishnan, Solutions Architect at Red Hat, discussed the impact commodity hardware, ubiquitous connectivity, and innovations in open source software are having on the connected universe of people, thi...
WebRTC: together these advances have created a perfect storm of technologies that are disrupting and transforming classic communications models and ecosystems. In his session at WebRTC Summit, Cary Bran, VP of Innovation and New Ventures at Plantronics and PLT Labs, provided an overview of this technological shift, including associated business and consumer communications impacts, and opportunities it may enable, complement or entirely transform.
There are so many tools and techniques for data analytics that even for a data scientist the choices, possible systems, and even the types of data can be daunting. In his session at @ThingsExpo, Chris Harrold, Global CTO for Big Data Solutions for EMC Corporation, showed how to perform a simple, but meaningful analysis of social sentiment data using freely available tools that take only minutes to download and install. Participants received the download information, scripts, and complete end-t...
For manufacturers, the Internet of Things (IoT) represents a jumping-off point for innovation, jobs, and revenue creation. But to adequately seize the opportunity, manufacturers must design devices that are interconnected, can continually sense their environment and process huge amounts of data. As a first step, manufacturers must embrace a new product development ecosystem in order to support these products.
Manufacturing connected IoT versions of traditional products requires more than multiple deep technology skills. It also requires a shift in mindset, to realize that connected, sensor-enabled “things” act more like services than what we usually think of as products. In his session at @ThingsExpo, David Friedman, CEO and co-founder of Ayla Networks, discussed how when sensors start generating detailed real-world data about products and how they’re being used, smart manufacturers can use the dat...
When it comes to IoT in the enterprise, namely the commercial building and hospitality markets, a benefit not getting the attention it deserves is energy efficiency, and IoT’s direct impact on a cleaner, greener environment when installed in smart buildings. Until now clean technology was offered piecemeal and led with point solutions that require significant systems integration to orchestrate and deploy. There didn't exist a 'top down' approach that can manage and monitor the way a Smart Buildi...