Click here to close now.

Welcome!

MICROSERVICES Authors: David Sprott, Michael Kanasoot, Carmen Gonzalez, Lori MacVittie, Elizabeth White

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We recognize that the average no longer represents the bulk of the transactions but can be a lot higher than the median.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

@MicroservicesExpo Stories
Right off the bat, Newman advises that we should "think of microservices as a specific approach for SOA in the same way that XP or Scrum are specific approaches for Agile Software development". These analogies are very interesting because my expectation was that microservices is a pattern. So I might infer that microservices is a set of process techniques as opposed to an architectural approach. Yet in the book, Newman clearly includes some elements of concept model and architecture as well as p...
Cloud computing is changing the way we look at IT costs, according to industry experts on a recent Cloud Luminary Fireside Chat panel discussion. Enterprise IT, traditionally viewed as a cost center, now plays a central role in the delivery of software-driven goods and services. Therefore, companies need to understand their cloud utilization and resulting costs in order to ensure profitability on their business offerings. Led by Bernard Golden, this fireside chat offers valuable insights on ho...
Microservices, for the uninitiated, are essentially the decomposition of applications into multiple services. This decomposition is often based on functional lines, with related functions being grouped together into a service. While this may sound a like SOA, it really isn't, especially given that SOA was an object-centered methodology that focused on creating services around "nouns" like customer and product. Microservices, while certainly capable of being noun-based, are just as likely to be v...
SYS-CON Events announced today the DevOps Foundation Certification Course, being held June ?, 2015, in conjunction with DevOps Summit and 16th Cloud Expo at the Javits Center in New York City, NY. This sixteen (16) hour course provides an introduction to DevOps – the cultural and professional movement that stresses communication, collaboration, integration and automation in order to improve the flow of work between software developers and IT operations professionals. Improved workflows will res...
Chef and Canonical announced a partnership to integrate and distribute Chef with Ubuntu. Canonical is integrating the Chef automation platform with Canonical's Machine-As-A-Service (MAAS), enabling users to automate the provisioning, configuration and deployment of bare metal compute resources in the data center. Canonical is packaging Chef 12 server in upcoming distributions of its Ubuntu open source operating system and will provide commercial support for Chef within its user base.
You hear the terms “subscription economy” and “subscription commerce” all the time. And with good reason. Subscription-based monetization is transforming business as we know it. But what about usage? Where’s the “consumption economy”? Turns out, it’s all around us. When most people think of usage-based billing, the example that probably comes to mind first is metered public utilities — water, gas and electric. Phone services, especially mobile, might come next. Then maybe taxis. And that’s ab...
Learn the top API testing issues that organizations encounter and how automation plus a DevOps team approach can address these top API testing challenges. Ensuring API integrity is difficult in today's complex application cloud, on-premises and hybrid environment scenarios. In this interview with TechTarget, Parasoft solution architect manager Spencer Debrosse shares his experiences about the top API testing issues that organizations encounter and how automation and a DevOps team approach can a...
Containers and microservices have become topics of intense interest throughout the cloud developer and enterprise IT communities. Accordingly, attendees at the upcoming 16th Cloud Expo at the Javits Center in New York June 9-11 will find fresh new content in a new track called PaaS | Containers & Microservices Containers are not being considered for the first time by the cloud community, but a current era of re-consideration has pushed them to the top of the cloud agenda. With the launch ...
An explosive combination of technology trends will be where ‘microservices’ and the IoT Internet of Things intersect, a concept we can describe by comparing it with a previous theme, the ‘X Internet.' The idea of using small self-contained application components has been popular since XML Web services began and a distributed computing future of smart fridges and kettles was imagined long back in the early Internet years.
After what feel like an interminable cycle of media frenzy followed by hype and hysteria cycles, the practical elements of real world cloud implementations are starting to become better documented. But what is really different in the cloud? How do software applications behave, live, interact and interconnect inside the cloud? Where do cloud architectures differ so markedly from their predecessors that we need to learn a new set of mechanics – and, when do we start to refer to software progra...
SOA Software has changed its name to Akana. With roots in Web Services and SOA Governance, Akana has established itself as a leader in API Management and is expanding into cloud integration as an alternative to the traditional heavyweight enterprise service bus (ESB). The company recently announced that it achieved more than 90% year-over-year growth. As Akana, the company now addresses the evolution and diversification of SOA, unifying security, management, and DevOps across SOA, APIs, microser...
It's 2:15pm on a Friday, and I'm sitting in the keynote hall at PyCon 2013 fidgeting through a succession of lightning talks that have very little relevance to my life. Topics like "Python code coverage techniques" (ho-hum) and "Controlling Christmas lights with Python” (yawn - I wonder if there's anything new on Hacker News)...when Solomon Hykes takes the stage, unveils Docker, and the world shifts. If you haven't seen it yet, you should watch the video of Solomon's Pycon The Future of Linux C...
This month I want to revisit supporting infrastructure and datacenter environments. I have touched (some would say rant) upon this topic since my post in April 2014 called "Take a Holistic View of Support". My thoughts and views on this topic have not changed at all: it's critical for any organization to have a holistic, comprehensive strategy and view of how they support their IT infrastructure and datacenter environments. In fact, I believe it's even more critical today then it was a year ago ...
The 16th Cloud Expo has added coverage containers and microservices to its program for New York, to be held June 9-11 at the Jacob Javits Convention Center. Cloud Expo has long been the single, independent show where delegates and technology vendors can meet to experience and discuss the entire world of the cloud. This year will be no different. Containers are an old concept that saw renewed life with the emergence of Docker in 2013. Then late in 2014, CoreOS shook up the cloud-computing w...
OmniTI has expanded its services to help customers automate their processes to deliver high quality applications to market faster. Consistent with its focus on IT agility and quality, OmniTI operates under DevOps principles, exploring the flow of value through the IT delivery process, identifying opportunities to eliminate waste, realign misaligned incentives, and open bottlenecks. OmniTI takes a unique, value-centric approach by plotting each opportunity in an effort-payoff quadrant, then work...
SYS-CON Events announced today that Solgenia will exhibit at SYS-CON's 16th International Cloud Expo®, which will take place on June 9-11, 2015, at the Javits Center in New York City, NY, and the 17th International Cloud Expo®, which will take place on November 3–5, 2015, at the Santa Clara Convention Center in Santa Clara, CA. Solgenia is the global market leader in Cloud Collaboration and Cloud Infrastructure software solutions. Designed to “Bridge the Gap” between Personal and Professional S...
SYS-CON Events announced today that MangoApps will exhibit at SYS-CON's 16th International Cloud Expo®, which will take place on June 9-11, 2015, at the Javits Center in New York City, NY., and the 17th International Cloud Expo®, which will take place on November 3–5, 2015, at the Santa Clara Convention Center in Santa Clara, CA. MangoApps provides private all-in-one social intranets allowing workers to securely collaborate from anywhere in the world and from any device. Social, mobile, and eas...
When it comes to microservices there are myths and uncertainty about the journey ahead. Deploying a “Hello World” app on Docker is a long way from making microservices work in real enterprises with large applications, complex environments and existing organizational structures. February 19, 2015 10:00am PT / 1:00pm ET → 45 Minutes Join our four experts: Special host Gene Kim, Gary Gruver, Randy Shoup and XebiaLabs’ Andrew Phillips as they explore the realities of microservices in today’s IT worl...
The world's leading Cloud event, Cloud Expo has launched Microservices Journal on the SYS-CON.com portal, featuring over 19,000 original articles, news stories, features, and blog entries. DevOps Journal is focused on this critical enterprise IT topic in the world of cloud computing. Microservices Journal offers top articles, news stories, and blog posts from the world's well-known experts and guarantees better exposure for its authors than any other publication. Follow new article posts on T...
Hosted PaaS providers have given independent developers and startups huge advantages in efficiency and reduced time-to-market over their more process-bound counterparts in enterprises. Software frameworks are now available that allow enterprise IT departments to provide these same advantages for developers in their own organization. In his workshop session at DevOps Summit, Troy Topnik, ActiveState’s Technical Product Manager, will show how on-prem or cloud-hosted Private PaaS can enable organ...