|By Michael Kopp||
|December 12, 2012 07:30 AM EST||
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We recognize that the average no longer represents the bulk of the transactions but can be a lot higher than the median.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.
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!
|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?
Application transformation and DevOps practices are two sides of the same coin. Enterprises that want to capture value faster, need to deliver value faster – time value of money principle. To do that enterprises need to build cloud-native apps as microservices by empowering teams to build, ship, and run in production. In his session at @DevOpsSummit at 19th Cloud Expo, Neil Gehani, senior product manager at HPE, discussed what every business should plan for how to structure their teams to delive...
Dec. 3, 2016 08:30 AM EST Reads: 1,343
Rapid innovation, changing business landscapes, and new IT demands force businesses to make changes quickly. In the eyes of many, containers are at the brink of becoming a pervasive technology in enterprise IT to accelerate application delivery. In this presentation, attendees learned about the: The transformation of IT to a DevOps, microservices, and container-based architecture What are containers and how DevOps practices can operate in a container-based environment A demonstration of how ...
Dec. 3, 2016 08:30 AM EST Reads: 658
As we enter the final week before the 19th International Cloud Expo | @ThingsExpo in Santa Clara, CA, it's time for me to reflect on six big topics that will be important during the show. Hybrid Cloud This general-purpose term seems to provide a comfort zone for many enterprise IT managers. It sounds reassuring to be able to work with one of the major public-cloud providers like AWS or Microsoft Azure while still maintaining an on-site presence.
Dec. 3, 2016 04:00 AM EST Reads: 2,701
Without lifecycle traceability and visibility across the tool chain, stakeholders from Planning-to-Ops have limited insight and answers to who, what, when, why and how across the DevOps lifecycle. This impacts the ability to deliver high quality software at the needed velocity to drive positive business outcomes. In his general session at @DevOpsSummit at 19th Cloud Expo, Phil Hombledal, Solution Architect at CollabNet, discussed how customers are able to achieve a level of transparency that e...
Dec. 3, 2016 02:15 AM EST Reads: 747
Much of the value of DevOps comes from a (renewed) focus on measurement, sharing, and continuous feedback loops. In increasingly complex DevOps workflows and environments, and especially in larger, regulated, or more crystallized organizations, these core concepts become even more critical. In his session at @DevOpsSummit at 18th Cloud Expo, Andi Mann, Chief Technology Advocate at Splunk, showed how, by focusing on 'metrics that matter,' you can provide objective, transparent, and meaningful f...
Dec. 3, 2016 01:45 AM EST Reads: 4,533
Between 2005 and 2020, data volumes will grow by a factor of 300 – enough data to stack CDs from the earth to the moon 162 times. This has come to be known as the ‘big data’ phenomenon. Unfortunately, traditional approaches to handling, storing and analyzing data aren’t adequate at this scale: they’re too costly, slow and physically cumbersome to keep up. Fortunately, in response a new breed of technology has emerged that is cheaper, faster and more scalable. Yet, in meeting these new needs they...
Dec. 3, 2016 12:15 AM EST Reads: 1,754
@DevOpsSummit taking place June 6-8, 2017 at Javits Center, New York City, is co-located with the 20th International Cloud Expo and will feature technical sessions from a rock star conference faculty and the leading industry players in the world. @DevOpsSummit at Cloud Expo New York Call for Papers is now open.
Dec. 2, 2016 10:30 PM EST Reads: 1,734
Logs are continuous digital records of events generated by all components of your software stack – and they’re everywhere – your networks, servers, applications, containers and cloud infrastructure just to name a few. The data logs provide are like an X-ray for your IT infrastructure. Without logs, this lack of visibility creates operational challenges for managing modern applications that drive today’s digital businesses.
Dec. 2, 2016 08:00 PM EST Reads: 1,541
Today we can collect lots and lots of performance data. We build beautiful dashboards and even have fancy query languages to access and transform the data. Still performance data is a secret language only a couple of people understand. The more business becomes digital the more stakeholders are interested in this data including how it relates to business. Some of these people have never used a monitoring tool before. They have a question on their mind like “How is my application doing” but no id...
Dec. 2, 2016 04:45 PM EST Reads: 2,113
Information technology is an industry that has always experienced change, and the dramatic change sweeping across the industry today could not be truthfully described as the first time we've seen such widespread change impacting customer investments. However, the rate of the change, and the potential outcomes from today's digital transformation has the distinct potential to separate the industry into two camps: Organizations that see the change coming, embrace it, and successful leverage it; and...
Dec. 2, 2016 03:30 PM EST Reads: 3,208
In IT, we sometimes coin terms for things before we know exactly what they are and how they’ll be used. The resulting terms may capture a common set of aspirations and goals – as “cloud” did broadly for on-demand, self-service, and flexible computing. But such a term can also lump together diverse and even competing practices, technologies, and priorities to the point where important distinctions are glossed over and lost.
Dec. 2, 2016 03:15 PM EST Reads: 1,454
Monitoring of Docker environments is challenging. Why? Because each container typically runs a single process, has its own environment, utilizes virtual networks, or has various methods of managing storage. Traditional monitoring solutions take metrics from each server and applications they run. These servers and applications running on them are typically very static, with very long uptimes. Docker deployments are different: a set of containers may run many applications, all sharing the resource...
Dec. 2, 2016 01:45 PM EST Reads: 5,450
Keeping pace with advancements in software delivery processes and tooling is taxing even for the most proficient organizations. Point tools, platforms, open source and the increasing adoption of private and public cloud services requires strong engineering rigor – all in the face of developer demands to use the tools of choice. As Agile has settled in as a mainstream practice, now DevOps has emerged as the next wave to improve software delivery speed and output. To make DevOps work, organization...
Dec. 2, 2016 01:30 PM EST Reads: 5,704
Join Impiger for their featured webinar: ‘Cloud Computing: A Roadmap to Modern Software Delivery’ on November 10, 2016, at 12:00 pm CST. Very few companies have not experienced some impact to their IT delivery due to the evolution of cloud computing. This webinar is not about deciding whether you should entertain moving some or all of your IT to the cloud, but rather, a detailed look under the hood to help IT professionals understand how cloud adoption has evolved and what trends will impact th...
Dec. 2, 2016 01:00 PM EST Reads: 2,453
Internet of @ThingsExpo, taking place June 6-8, 2017 at the Javits Center in New York City, New York, is co-located with the 20th International Cloud Expo and will feature technical sessions from a rock star conference faculty and the leading industry players in the world. @ThingsExpo New York Call for Papers is now open.
Dec. 2, 2016 12:00 PM EST Reads: 1,843
Without lifecycle traceability and visibility across the tool chain, stakeholders from Planning-to-Ops have limited insight and answers to who, what, when, why and how across the DevOps lifecycle. This impacts the ability to deliver high quality software at the needed velocity to drive positive business outcomes. In his session at @DevOpsSummit 19th Cloud Expo, Eric Robertson, General Manager at CollabNet, showed how customers are able to achieve a level of transparency that enables everyone fro...
Dec. 2, 2016 11:30 AM EST Reads: 1,778
You have great SaaS business app ideas. You want to turn your idea quickly into a functional and engaging proof of concept. You need to be able to modify it to meet customers' needs, and you need to deliver a complete and secure SaaS application. How could you achieve all the above and yet avoid unforeseen IT requirements that add unnecessary cost and complexity? You also want your app to be responsive in any device at any time. In his session at 19th Cloud Expo, Mark Allen, General Manager of...
Dec. 2, 2016 10:45 AM EST Reads: 1,618
The 20th International Cloud Expo has announced that its Call for Papers is open. Cloud Expo, to be held June 6-8, 2017, at the Javits Center in New York City, brings together Cloud Computing, Big Data, Internet of Things, DevOps, Containers, Microservices and WebRTC to one location. With cloud computing driving a higher percentage of enterprise IT budgets every year, it becomes increasingly important to plant your flag in this fast-expanding business opportunity. Submit your speaking proposal ...
Dec. 2, 2016 10:15 AM EST Reads: 2,057
"Dice has been around for the last 20 years. We have been helping tech professionals find new jobs and career opportunities," explained Manish Dixit, VP of Product and Engineering at Dice, in this SYS-CON.tv interview at 19th Cloud Expo, held November 1-3, 2016, at the Santa Clara Convention Center in Santa Clara, CA.
Dec. 2, 2016 08:15 AM EST Reads: 810
In his session at 19th Cloud Expo, Claude Remillard, Principal Program Manager in Developer Division at Microsoft, contrasted how his team used config as code and immutable patterns for continuous delivery of microservices and apps to the cloud. He showed how the immutable patterns helps developers do away with most of the complexity of config as code-enabling scenarios such as rollback, zero downtime upgrades with far greater simplicity. He also demoed building immutable pipelines in the cloud ...
Dec. 1, 2016 09:00 PM EST Reads: 1,718