Welcome!

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We recognize that the average no longer represents the bulk of the transactions but can be a lot higher than the median.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

Latest Stories
SYS-CON Events announced today that Pulzze Systems will exhibit at the 19th International Cloud Expo, which will take place on November 1–3, 2016, at the Santa Clara Convention Center in Santa Clara, CA. Pulzze Systems, Inc. provides infrastructure products for the Internet of Things to enable any connected device and system to carry out matched operations without programming. For more information, visit http://www.pulzzesystems.com.
If you’re responsible for an application that depends on the data or functionality of various IoT endpoints – either sensors or devices – your brand reputation depends on the security, reliability, and compliance of its many integrated parts. If your application fails to deliver the expected business results, your customers and partners won't care if that failure stems from the code you developed or from a component that you integrated. What can you do to ensure that the endpoints work as expect...
DevOps at Cloud Expo – being held November 1-3, 2016, at the Santa Clara Convention Center in Santa Clara, CA – announces that its Call for Papers is open. Born out of proven success in agile development, cloud computing, and process automation, DevOps is a macro trend you cannot afford to miss. From showcase success stories from early adopters and web-scale businesses, DevOps is expanding to organizations of all sizes, including the world's largest enterprises – and delivering real results. Am...
The Internet of Things can drive efficiency for airlines and airports. In their session at @ThingsExpo, Shyam Varan Nath, Principal Architect with GE, and Sudip Majumder, senior director of development at Oracle, will discuss the technical details of the connected airline baggage and related social media solutions. These IoT applications will enhance travelers' journey experience and drive efficiency for the airlines and the airports. The session will include a working demo and a technical d...
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 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 solutions, helping companies worldwide activate their data to drive more value and business insight and to transform moder...
The Transparent Cloud-computing Consortium (abbreviation: T-Cloud Consortium) will conduct research activities into changes in the computing model as a result of collaboration between "device" and "cloud" and the creation of new value and markets through organic data processing High speed and high quality networks, and dramatic improvements in computer processing capabilities, have greatly changed the nature of applications and made the storing and processing of data on the network commonplace.
SYS-CON Events announced today that Bsquare has been named “Silver Sponsor” of SYS-CON's @ThingsExpo, which will take place on November 1–3, 2016, at the Santa Clara Convention Center in Santa Clara, CA. For more than two decades, Bsquare has helped its customers extract business value from a broad array of physical assets by making them intelligent, connecting them, and using the data they generate to optimize business processes.
I'm a lonely sensor. I spend all day telling the world how I'm feeling, but none of the other sensors seem to care. I want to be connected. I want to build relationships with other sensors to be more useful for my human. I want my human to understand that when my friends next door are too hot for a while, I'll soon be flaming. And when all my friends go outside without me, I may be left behind. Don't just log my data; use the relationship graph. In his session at @ThingsExpo, Ryan Boyd, Engi...
Fact is, enterprises have significant legacy voice infrastructure that’s costly to replace with pure IP solutions. How can we bring this analog infrastructure into our shiny new cloud applications? There are proven methods to bind both legacy voice applications and traditional PSTN audio into cloud-based applications and services at a carrier scale. Some of the most successful implementations leverage WebRTC, WebSockets, SIP and other open source technologies. In his session at @ThingsExpo, Da...
SYS-CON Events announced today that Secure Channels will exhibit at the 19th International Cloud Expo, which will take place on November 1–3, 2016, at the Santa Clara Convention Center in Santa Clara, CA. The bedrock of Secure Channels Technology is a uniquely modified and enhanced process based on superencipherment. Superencipherment is the process of encrypting an already encrypted message one or more times, either using the same or a different algorithm.
Traditional on-premises data centers have long been the domain of modern data platforms like Apache Hadoop, meaning companies who build their business on public cloud were challenged to run Big Data processing and analytics at scale. But recent advancements in Hadoop performance, security, and most importantly cloud-native integrations, are giving organizations the ability to truly gain value from all their data. In his session at 19th Cloud Expo, David Tishgart, Director of Product Marketing ...
Almost two-thirds of companies either have or soon will have IoT as the backbone of their business in 2016. However, IoT is far more complex than most firms expected. How can you not get trapped in the pitfalls? In his session at @ThingsExpo, Tony Shan, a renowned visionary and thought leader, will introduce a holistic method of IoTification, which is the process of IoTifying the existing technology and business models to adopt and leverage IoT. He will drill down to the components in this fra...
The vision of a connected smart home is becoming reality with the application of integrated wireless technologies in devices and appliances. The use of standardized and TCP/IP networked wireless technologies in line-powered and battery operated sensors and controls has led to the adoption of radios in the 2.4GHz band, including Wi-Fi, BT/BLE and 802.15.4 applied ZigBee and Thread. This is driving the need for robust wireless coexistence for multiple radios to ensure throughput performance and th...
Enterprise IT has been in the era of Hybrid Cloud for some time now. But it seems most conversations about Hybrid are focused on integrating AWS, Microsoft Azure, or Google ECM into existing on-premises systems. Where is all the Private Cloud? What do technology providers need to do to make their offerings more compelling? How should enterprise IT executives and buyers define their focus, needs, and roadmap, and communicate that clearly to the providers?
SYS-CON Events announced today the Kubernetes and Google Container Engine Workshop, being held November 3, 2016, in conjunction with @DevOpsSummit at 19th Cloud Expo at the Santa Clara Convention Center in Santa Clara, CA. This workshop led by Sebastian Scheele introduces participants to Kubernetes and Google Container Engine (GKE). Through a combination of instructor-led presentations, demonstrations, and hands-on labs, students learn the key concepts and practices for deploying and maintainin...