Welcome!

Cloud Expo Authors: Elizabeth White, Pat Romanski, Trevor Parsons, Lori MacVittie, Cynthia Dunlop

Related Topics: Java, SOA & WOA, Adobe Flex, AJAX & REA, Apache

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We recognize that the average no longer represents the bulk of the transactions but can be a lot higher than the median.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

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.


@CloudExpo Stories
SYS-CON Events announced today that SOA Software, an API management leader, will exhibit at SYS-CON's 15th International Cloud Expo®, which will take place on November 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA. SOA Software is a leading provider of API Management and SOA Governance products that equip business to deliver APIs and SOA together to drive their company to meet its business strategy quickly and effectively. SOA Software’s technology helps businesses to accel...
As cloud gives an opportunity to businesses to buy services externally - how is cloud impacting your customers? In his General Session at 15th Cloud Expo, Fabio Gori, Director of Worldwide Cloud Marketing at Cisco, will provide answers to big questions: Do you see hybrid cloud as where the world is going? What benefits does it bring? And how does Cisco connect all of these clouds? He will also tell us everything about Intercloud and Cisco investment on it.
Can we look to the paradigm of cloud computing from a completely different perspective? In his General Session at 15th Cloud Expo, Gundars Kulups, Sales Director at DEAC, will discuss what we can learn from our dining habits when choosing a cloud solution. Gundars Kulups is Sales Director at DEAC, full service data center operator. An IT expert, he specializes in European countries and has worked in the IT industry since 1992. He graduated from Riga Technical University (RTU) in Latvia and com...
SYS-CON Events announced today that TMCnet has been named “Media Sponsor” of SYS-CON's 15th International Cloud Expo®, which will take place on November 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA. Technology Marketing Corporation (TMC) is the world's leading business to business and integrated marketing media company, servicing niche markets within the communications and technology industries.
SYS-CON Events announced today that Aria Systems, the recurring revenue expert, has been named "Bronze Sponsor" of SYS-CON's 15th International Cloud Expo®, which will take place on November 4-6, 2014, at the Santa Clara Convention Center in Santa Clara, CA. Aria Systems helps leading businesses connect their customers with the products and services they love. Industry leaders like Pitney Bowes, Experian, AAA NCNU, VMware, HootSuite and many others choose Aria to power their recurring revenue bu...
SYS-CON Events announced today that Parasoft will exhibit at SYS-CON's 15th International Cloud Expo®, which will take place on November 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA. For 27 years, Parasoft has researched and developed software solutions that help organizations deliver defect-free software efficiently. By integrating Development Testing, API/cloud/SOA/composite app testing, and service virtualization, we reduce the time, effort, and cost of delivering secur...
The Internet of Things (IoT) promises to evolve the way the world does business; however, understanding how to apply it to your company can be a mystery. Most people struggle with understanding the potential business uses or tend to get caught up in the technology, resulting in solutions that fail to meet even minimum business goals. In his session at Internet of @ThingsExpo, Jesse Shiah, CEO / President / Co-Founder of AgilePoint Inc., will show what is needed to leverage the IoT to transform...
SYS-CON Events announced today that Utimaco will exhibit at SYS-CON's 15th International Cloud Expo®, which will take place on November 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA. Utimaco is a leading manufacturer of hardware based security solutions that provide the root of trust to keep cryptographic keys safe, secure critical digital infrastructures and protect high value data assets. Only Utimaco delivers a general-purpose hardware security module (HSM) as a customiz...
One of the biggest challenges when developing connected devices is identifying user value and delivering it through successful user experiences. In his session at Internet of @ThingsExpo, Mike Kuniavsky, Principal Scientist, Innovation Services at PARC, will describe an IoT-specific approach to user experience design that combines approaches from interaction design, industrial design and service design to create experiences that go beyond simple connected gadgets to create lasting, multi-devic...
Until recently, many organizations required specialized departments to perform mapping and geospatial analysis, and they used Esri on-premise solutions for that work. In his session at 15th Cloud Expo, Dave Peters, author of the Esri Press book Building a GIS, System Architecture Design Strategies for Managers, will discuss how Esri has successfully included the cloud as a fully integrated SaaS expansion of the ArcGIS mapping platform. Organizations that have incorporated Esri cloud-based appl...
Dyn solutions are at the core of Internet Performance. Through traffic management, message management and performance assurance, Dyn is connecting people through the Internet and ensuring information gets where it needs to go, faster and more reliably than ever before. Founded in 2001 at WPI, Dyn’s global presence services more than four million enterprise, small business and personal customers.
Once the decision has been made to move part or all of a workload to the cloud, a methodology for selecting that workload needs to be established. How do you move to the cloud? What does the discovery, assessment and planning look like? What workloads make sense? Which cloud model makes sense for each workload? What are the considerations for how to select the right cloud model? And how does that fit in with the overall IT tranformation? In his session at 15th Cloud Expo, John Hatem, head of V...
IBM and Tencent Cloud signed a business cooperation memorandum to collaborate on providing public cloud with Software-as-a-Service solutions for industries. Both parties agreed to focus on emerging small and medium enterprises in the smarter cities and smarter healthcare industries as well as other fields. This will enable these industries to utilize mobile, cloud computing and big data tools to transform internal processes and operations, thus achieving cloud transformation in the era of mobili...
SimpleECM is the only platform to offer a powerful combination of enterprise content management (ECM) services, capture solutions, and third-party business services providing simplified integrations and workflow development for solution providers. SimpleECM is opening the market to businesses of all sizes by reinventing the delivery of ECM services. Our APIs make the development of ECM services simple with the use of familiar technologies for a frictionless integration directly into web applicat...
European data center operator DEAC is the largest in the Baltics. The activities are orientated to provide data center services and IT outsourcing on Eurasia and America scale in order to create the primary or backup or additional data center for customer in the EU, to protect its business and, most importantly, reduce costs up to 40% within 3-5 years. DEAC is an IT outsourcing services and solutions company whose highly experienced and qualified employees offer various groups of services and...
The Internet of Things will greatly expand the opportunities for data collection and new business models driven off of that data. In her session at Internet of @ThingsExpo, Esmeralda Swartz, CMO of MetraTech, will discuss how for this to be effective you not only need to have infrastructure and operational models capable of utilizing this new phenomenon, but increasingly service providers will need to convince a skeptical public to participate. Get ready to show them the money! Speaker Bio: ...
Samsung VP Jacopo Lenzi, who headed the company's recent SmartThings acquisition under the auspices of Samsung's Open Innovaction Center (OIC), answered a few questions we had about the deal. This interview was in conjunction with our interview with SmartThings CEO Alex Hawkinson. IoT Journal: SmartThings was developed in an open, standards-agnostic platform, and will now be part of Samsung's Open Innovation Center. Can you elaborate on your commitment to keep the platform open? Jacopo Lenzi: S...
The major cloud platforms defy a simple, side-by-side analysis. Each of the major IaaS public-cloud platforms offers their own unique strengths and functionality. Options for on-site private cloud are diverse as well, and must be designed and deployed while taking existing legacy architecture and infrastructure into account. Then the reality is that most enterprises are embarking on a hybrid cloud strategy and programs. In this Power Panel at 15th Cloud Expo, moderated by Ashar Baig, Research ...
Things are being built upon cloud foundations to transform organizations. This CEO Power Panel at 15th Cloud Expo, moderated by Roger Strukhoff, Cloud Expo and @ThingsExpo conference chair, will address the big issues involving these technologies and, more important, the results they will achieve. How important are public, private, and hybrid cloud to the enterprise? How does one define Big Data? And how is the IoT tying all this together?
When an enterprise builds a hybrid IaaS cloud connecting its data center to one or more public clouds, security is often a major topic along with the other challenges involved. Security is closely intertwined with the networking choices made for the hybrid cloud. Traditional networking approaches for building a hybrid cloud try to kludge together the enterprise infrastructure with the public cloud. Consequently this approach requires risky, deep "surgery" including changes to firewalls, subnets...