Akamai Diversity
Home > Web Performance

Recently in Web Performance Category

The holiday season is already creeping up and by far will be the most vital online shopping period of the year for retailers. Thanksgiving week will no doubt once again present one of the largest online shopping weeks of the year.  Now, more than ever, time is literally money when it comes to Web performance.  

 

As online retailers face another big holiday shopping season, they have to make sure they are ready for everything and anything that will hit their site once the critical holiday season begins.  Even a minute of downtime can cost thousands and thousands of dollars and can damage the bottom line and the brand for years to come.


Supporting the majority of today's leading global retailers, Akamai has the expertise and technology to help retailers simplify the process of preparing their e-commerce site for Black Friday as well as helping them to deliver on every big shopping day before and after with situational performance - so that any customer, anywhere can make a purchase as soon as inspiration strikes.


Starting this week, Akamai will be launching a series of "Crush The Rush" Holiday Readiness Webinars to help retailers understand how to properly prepare their site for the biggest shopping season to-date.  Participants will learn first hand how leading retailers have deployed Akamai Solutions specially tuned for holiday traffic, ensuring their sites are scalable and reliable as customers flood their online store.


The "Crush the Rush" series consists of three Webinars focused on different topics relevant to preparing for the rush of the online holiday season.  The first one will be held on July 24th on Situational Performance.  Participants will come away with clear understanding of why looking at performance from an end-user perspective is key for those who wish to "Crush the Rush" this holiday season.  Our speaker, Lorenz Jakober, will be discussing what is Real-User Monitoring (RUM), synthetic vs. RUM testing and the pros and cons for each, why optimizing beyond average load times matters and how performance optimization helps meet rising consumer expectations.


Following Situational Performance, we will also be discussing how Akamai Services & Support engage with the industry's leading brands as well as Web Security and Risk Mitigation throughout the Webinar series.  Our speakers will be sharing recaps and best practices that retailers should be aware of for this holiday season.  Follow us here to find out more details on the "Crush the Rush" Holiday Readiness Webinar Series.  We want to help you beat the rush to the holiday rush.


The DDoS Paradox

According to the Department of Homeland Security, almost 50 US Financial Institutions have suffered more than 200 Distributed Denial of Service attacks since September 2012 . Because we protect the majority of world's biggest banks, asset management firms, and online brokers, Akamai is in the unique position of having witnessed and actively defended against many of these attacks, and can describe the evolution of attack targets as well as attack techniques. 
Over the past few months, we've seen attackers migrate towards two broad techniques:

  1. Request large objects (PDFs, image files, etc.)
  2. Attack non-cacheable pages (login pages, pages served by adwords, etc.)

Security professionals will be neither surprised nor impressed by these findings. Nor will they question that unprotected sites typically suffer increased response times or downtime when they are victim to these attacks. What might surprise them, however, is how the common responses to these threats is leading, in some cases, to increased latency in sites even when they are not under attack, and in some cases are leaving sites more likely to crash or suffer data exfiltration than they were before "preparedness steps" were taken.

The DDoS Paradox
The tendency to tighten rules and broaden inspections to the point of decreasing performance is what we have come to describe as the "DDoS Paradox". The logical thinking that leads to the paradox is as follows:

  1. CSO at Company ABC reads about attacks.
  2. CSO tightens and broadens rules on Web Application Firewall in order to better prepare for attacks.
  3. Tightened and broadened rules lead to increased inspection of incoming requests which slows down legitimate traffic and makes it easier for malicious traffic to flood and knock down the WAF.

The first outcome (slowing down legitimate traffic) is clearly bad for Company ABC, and good for the threat actors who are looking to cause widespread interruptions to economic activity. The second outcome (knocking down the WAF) is unfortunately good for threat actors who are trying to steal data. If they've launched an application layer DDoS attack that knocks down a firewall, they can then move in with a relatively simple SQLi or XSS attack in order to steal data or install malware on site visitors' PCs.  

For companies trying to protect their web assets, the DDoS Paradox presents a lose/lose situation. Fortunately, there are ways around the paradox. Interestingly, these options involve tightening and broadening WAF rules outside of the data center. In other words, tightening and broadening rules at the edge of the Internet is the best way to ensure that your tighter security measures do not inadvertently lead to deprecation in performance and/or an increased susceptibility to data theft.


Akamai's Kona Security Solutions do just that --- they provide inline, always on, and highly scalable DDoS and application layer defense at the edge of the internet, giving CSOs the ability to respond to attacks without suffering trade offs.

 

Dan Shugrue is a senior product marketing manager at Akamai.

It has been a year since I last wrote about World IPv6 Launch and our measurements of IPv6 adoption at the time. Since then, we've seen continued momentum around increased IPv6 adoption on multiple fronts, with more IPv6 end users as well as with more content becoming available over IPv6. The net result of this is that IPv6 traffic on Akamai's global platform has increased to be over 250% of its June 2012 level, and we are now delivering around 10 billion requests per day over IPv6, up from around 3.4 billion requests per day at this time last year. Over the course of a given week, Akamai is now seeing between 200 million and 300 million unique IPv6 addresses contact our network.

In addition, we have observed continued momentum in IPv6 adoption by our customers, with over twice as many customers delivering IPv6-enabled properties from our platform as we had at World IPv6 Launch last June.

Akamai also continues to increase the footprint of our IPv6 network deployment as more of our network partners make IPv6 connectivity available. We now have IPv6 enabled in 64 countries and over 800 network locations around the globe.

IPV6_Image1.jpg
IPv6 Requests/Day on Akamai from June 2012 to June 2013

While IPv4 is still the dominant protocol on the Internet and will be for years to come, IPv6 adoption continues to move forward, especially in the mobile space and in some parts of the world. We explore this in more detail below.

clickfrenzy.jpeg

Last year's Click Frenzy online sale in Australia, modeled after the hugely successful Black Friday sale in the US, attracted a huge amount of media attention for all the wrong reasons. Within minutes of the sale going live on November 20 last year, the website buckled under the strain of unanticipated traffic volumes. A number of consumers, excited at the prospect of grabbing a great online bargain, were left empty handed and disappointed.  The media's reaction was swift, and merciless. 

Fast-forward six months and the latest Click Frenzy sale was completed without any technical issues at all. Click Frenzy needed an industry leading solution, and as such enlisted Akamai.

Following their initial sale and the subsequent technical issues they encountered, Click Frenzy wanted to understand how they could handle the sudden bursts in traffic volumes that characterizes their online business model.  For Click Frenzy, it was absolutely imperative that their next big sale performed seamlessly to restore consumer and retailer confidence.  Further technical issues had to be avoided at all costs. 

We approached Click Frenzy and explained how Akamai's intelligent platform - specifically Akamai's AQUA & KONA solutions - enables traffic to be securely offloaded to distributed computing resources, thereby alleviating the burden on the their data centre.  This model allows businesses, such as Click Frenzy, to manage traffic spikes that would be unheard of for most conventional online retailers. 

Following Click Frenzy's 24-hour sale last month, the Akamai platform managed 169 million requests, with a peak of 29,722 requests per second. Total traffic volume exceeded 3TB.  This article, which features an interview I did following the sale, provides a great overview of some of the positive outcomes Click Frenzy has enjoyed since its rollout of the Akamai intelligent platform.

Although the Click Frenzy site itself performed flawlessly, some technical issues were still evident on the part of some retailers involved in the promotion since their websites - not utilizing Akamai's platform - were unable to cope with the traffic being directed at them by Click Frenzy.

For retailers, having too many customers can be a nice problem to have, but customers can be unforgiving if they are running into the same issues time and again.  This is particularly true for online where a customer can 'enter' another store with a simple click of a mouse. 

A robust technology platform that enables customers to access the information they need and purchase in an efficient manner is absolutely essential in today's super-competitive online retail environment.  And retailers don't want to be bogged down by technology. They just want it to work, so they can focus on their core business and what they know best - retailing.  For our friends at Click Frenzy, that's exactly what they're doing now and we look forward to many more successful online 'frenzies', and maybe picking up a bargain or two ourselves along the way. 

 

Ian Teague is a regional sales manager at Akamai.

As we know, many of the world's leading commerce companies trust Akamai.  What about other mid-sized and smaller businesses and online retailers who are looking for acceleration and improvement of their site performance?  How is Akamai helping them succeed in their online business?  


Legendary Whitetails, a customer through our partner MICROS Retail, is a great success story in showcasing how a family-owned retailer offering unique causal apparel, gifts and gear for whitetail hunters can boost site performance, increase revenue and brand engagement and support its largest traffic day by leveraging Akamai's Aqua Web Solutions.


With no physical locations, Legendary Whitetails relies on its Web site as its primary storefront.  They are always seeking ways to improve the site experience because they know this leaves a lasting impression on their shoppers.  The retailer couldn't afford to disappoint their shoppers or lose sales due to a sluggish site.  They also needed to flawlessly handle traffic spikes associated with holiday shopping and promotions.  As a result, they went with Akamai.


Legendary Whitetails is using Akamai to deliver all of its site dynamic content and images.  Since implementing the Akamai solutions, the retailer has realized tremendous benefits.  It has offloaded 72% of its dynamic content to the Akamai platform, resulting in overall site performance improvement of nearly 200% on average.  It has also accelerated the delivery of dynamic content by 180%.  In fact, these improvements have made it possible for Legendary Whitetails to support its largest traffic day - 40,000 site visitors and over 650,000 page views - on Cyber Monday.


According to Ryan Johnson, eCommerce Manager at Legendary Whitetails, "Akamai is a key contributor in allowing us to grow 20% and achieve record sales in 2012.  We are also very encouraged that the solution will provide us with the capacity to maintain similar growth rates for the years to come."


To learn more about this customer case study, please click here.  By leveraging the Akamai platform, Legendary Whitetails is enabled to deliver content at scale, improve conversion rates and drive sales revenues.


Helen Yang is a public relations manager at Akamai

Akamai IO - Going Global

I'm writing this short post to update you on a significant change in Akamai IO's data source. Since its inception, IO's data was based on many websites, but most of those sites catered to a US audience. This means that the data set was biased - it included global traffic, but held a disproportionate amount of traffic from the US.

Starting February 16th, we've started using a new data stream, based on traffic from most Akamai customers - meaning a much more global distribution, enough to take away the entire bias. The new data set is also bigger, including over a billion requests each day, and uses a newer version of our device characterization engine.

Notable Changes
Looking at the data, you can see the market share has indeed changed.

For instance, the chart below shows IE's dominance fell from ~50% to ~40%, while Safari's share jumped to almost 10%. My guess is that this relates to the better sample even within North America, as opposed to the theory that Safari is just more popular outside the US.
Screen Shot 2013-03-14 at 3.02.21 PM.png

Another example is that Android's market share increased significantly to over 5% of the overall browsing traffic, and Opera's share more than doubled. Both browsers are known to be more popular outside the US, which is a likely explanation for this change.
Screen Shot 2013-03-14 at 3.02.54 PM.png

Then again, not everything changed. Chrome, Firefox and Mobile Safari maintained similar market shares, and even a global view doesn't help make Blackberry seem more dominant...

Screen Shot 2013-03-14 at 3.06.39 PM.png

This new global data set really highlights the need for a geographic split - showing the market shares by continent, country and perhaps even city. We're working on this feature, along with others, and hope to have it available by May.

Is Your Infrastructure Thinking Too Hard? (Part 1)

Akamai's Enterprise Architects regularly perform site assessments for customers to help maximize performance in terms of speed, scalability and reliability for their Akamai setup and their origin infrastructure. 

The first step in this process is what we call discovery, where we dig deep into the customer's web architecture and website. This allows us to understand the whole story of how the website functions, along with business goals and design decisions.

Out of all of the questions and answers during discovery, there's one particular question that consistently generates a large amount of conversation and high-value ideas. In fact, this question can be the launching point for a number of critical operational exercises like designing a new application, migrating to a new platform, troubleshooting or simply auditing: "How much of your infrastructure do you use when you serve a request?"

That is, how many of your databases, application servers, internal caches, load balancers and front-end servers have to spend time thinking about how to output the requested web page?

The answer varies depending on the request, but the important metric is "origin think time" or how long the origin (your servers) must think about a request once the request is made. If the page is cacheable and can be served directly by (say) Akamai at the edge of the network, the think time will be slim to nil. If it's not cacheable, like dynamic or uniquely-personalized content, your servers will spend a measurable amount of time processing the request.   

I recommend sitting down with a diagram of your web infrastructure and a spreadsheet, and asking the following questions:

•    What are the typical types of requests for non-cacheable data browsers make to my website? In a lot of cases, there might be only 5-6 general types of web pages on a typical site.
•    For each of those types, which and how many of my servers/processes need to process that request? e.g. 1 database server, 1 app server, 1 auth process (on app server).
•    How much wall-clock time does it take for my servers to process the request? Typically measured in seconds, and get numbers for both cold and warm browser caches, averaging multiple requests.
•    If you have the metrics, include a column for how many times a user might access that type of page during a typical session.

Here's an example of what a typical e-commerce site's think-time chart might look like:

Ringel Blog Image.png

This chart will be a good first step in identifying hot-spots in your infrastructure, and give you the insight you need to start doing triage on think-time issues.

In Part 2 of this article, I'll talk about some approaches you can use in your own web infrastructure to reduce origin think time.  

Matt Ringel is Enterprise Architect in Akamai's Professional Services team

Long gone are the days when you could gut-check website performance by running a straightforward synthetic backbone test on a simulated browser. The web has evolved, and the old ways of measuring performance do not line up with how end users interact with today's website.

The web -- and how websites perform in the real world -- are directly related to the various ways users interact with them. I covered this concept of situational performance and why it's important in an earlier post, "Web Performance: Why One size Doesn't fit all." A modern website has to work well on new browsers, old browsers, powerful desktop machines with great connectivity, and a long tail of mobile devices with a broad range of bandwidth situations.

If you care about performance, you have to measure it. If you want to measure it right, you absolutely must keep at least some of the different performance situations you expect to see in mind. Playing this out a bit, your once simple synthetic testing approach just got complex.

Flashlights and floodlights
A good synthetic monitoring solution is like a good flashlight: it gives you visibility to what you're pointing it at and that's just about it. The implication here is that if you want to know how your site performs for IE 9 users on a DSL connection, you can spin up a test and find out. Want to know your site performs for IE 9 users on a cable connection? That's a different test. IE 8? Different test. Chrome? Yup, more tests.

Modern synthetic testing tools like Keynote and Gomez have gotten very good at simulating the end user experience by measuring to the on load event with modern browsers on last mile networks. Despite these advances, the nature of synthetic testing means that monitoring even a small number of situations can quickly get out of control.

Following through with this analogy, if a synthetic monitoring solution is a flashlight, Real User Measurement (RUM for short) is a floodlight. With RUM, you can see how all of your users experience your content. Rather than having to pinpoint your important performance situations (and that's an entirely different can of worms on its own), you can aggregate across all of them and get a good sense of what real users are experiencing.

It is important to stress that RUM is not a replacement for synthetic testing, however. The two can, and should, be used side-by-side. Synthetic testing is key for identifying availability issues, performance issues like a slow third-party, or origin side issues like a misbehaving server.

The Case for Imagination

My daughter's favorite TV show is "SpongeBob SquarePants." She loves it. Nothing else even comes close. Because she's such a fan, I downloaded the top 100 episodes on the family iPad and we watched one in bed last Sunday.

In the episode, titular character SpongeBob and his pal Patrick claim that all you need to have fun is an empty cardboard box. When their friend Squidward asks why they're interested in just the box and not the TV it used to hold, SpongeBob replies, "We don't need television, as long as we have our...imagination."

spongebob.png

Even though he's just a cartoon character, SpongeBob's insights are very valuable. What if you applied his wisdom to your online marketing strategy? If you were more imaginative, could you make magic happen this holiday season?

Ask yourself these questions:
First off, do you use your imagination when it comes to cross-channel engagement?
 Because each form of media (display, video, mobile, social) is unique, you need to come up with an original approach for each in order to have a successful online marketing campaign. That is, every platform is a unique cardboard box. Play around with them individually and come up with something different. Your game plan for display isn't going to fit into your social media strategy. Your approaches, however, shouldn't be too disparate--remember to maintain a common theme across all channels.

Second, how inspired is your creative? 
I have seen far too many banners that are just plain bland, and their mobile counterparts are merely shrunken versions of them. Also, many in-stream video ads are simply repurposed from TV ads. We've already established that you need to tailor your strategy to the platform, so make sure your creative makes the most of each medium. Have some fun--that cardboard box could be a spaceship, and that other one could be a teleporter or time machine.

Third, do you think creatively when it comes to your data? 
I've written before about the need to use fresh and current data--in light of the explosion of technological platforms, it is crucial to know how to glean the most important information from the vast sea of available data. It takes some original thinking to determine which data sets are truly valuable and worthwhile to you. If you use the most relevant and enlightening data to inform your creative efforts, you're on the right track toward an innovative campaign.

Finally, are you thinking outside the box when you define your goals for success?
 If you've read my previous blog posts, I may sound like a broken record here: if you're stuck with the tunnel-vision view that last-click attribution is an accurate method of gauging marketing success, then you're not really seeing if your strategy is working or whether your dollars are well spent. You should stop zoning out in front of the TV--get off the couch and start using your mind. What about running campaigns to drive new site traffic instead of merely retargeting to your current ones? How about targeting lapsed users? To do so, it's time to align the right attribution model with your marketing goals. It's time to think up that interstellar spaceship.

I never thought I would learn anything from watching a cartoon with my daughter, but it makes sense that I could be inspired by a quirky children's show. Children and their beloved cartoon characters aren't burdened by rules, convention and routine. You don't need to be either. You have your imagination.

Avi Spivack is Product Marketing Strategist for Akamai

End-Users' Web Experience Expectations Just Keep Getting Higher

People have high expectations. Web performance is no different. End-users expect fast and engaging web experiences. It doesn't matter if they use a PC, smartphone or tablet to interact with your web application. According to recent end-user research 89 percent of tablet, 59 percent of smartphone and 83 percent of PC users expect a website to load in 3 seconds or less. Now if we compare web performance expectations to the latest aggregate web page load times across smartphones and PCs it's not a pretty picture.

In addition just like we expect every new iPhone to be better, and every new Ferrari to be faster, our expectations for web performance just keep getting higher. In 2011, 71% of mobile users expected web pages to be as fast or faster than on a desktop -- up from 58% in 2009.

Philip Tellis summed up this rise in expectations nicely during a recent Velocity conference: "What delighted users a few years ago is now an expected baseline, the absence of which will frustrate."

To find out if web application delivery professionals also perceived this rise in end-user expectations Akamai commissioned Forrester Consulting to survey 210 US IT professionals across various industries. The resulting report entitled "Shifting Performance Strategies And Solutions For Mobile And Web Delivery" can be found here.

This is the second (the first - Is Web and Mobile Application Performance A Business Issue? You bet it is.) in a series of posts where we will take a deeper look at the key findings from the report.

When we asked IT professionals whether end-user's expectations were higher than three years ago the resounding answer was yes. It seems like expectations are rising for both internal and external audiences. Enterprise users are no longer willing to suffer through subpar and slow applications that might not function across the latest browsers and mobile devices. In addition both internal and external audiences not only expect faster web applications, but also richer experiences that evolve more often.

  • 74 percent of external and 73 percent of internal end-users expect faster web applications

Sit Perf pt 2 image 1.png

  • 76 percent of external and 64 percent of internal end-users expect richer web applications

Site Perf pt 2 image 2.png

  • 61 percent of external and 60 percent of internal users expect web applications that are updated more frequently
Sit Perf pt 2 image 3.png

So what do these stats tell us?
External and internal users continue to expect richer and richer web applications that are constantly evolving. This usually equals more features, which equals more payload to deliver to the end-user's browser. But not only do end-users want richer more up to date experiences - they want apps that are fast - blazing fast.

In our next post we will cover why delivering quality experiences and blazing fast applications has become so difficult. In the meantime if you are interested in learning more download the report "Shifting Performance Strategies And Solutions For Mobile And Web Delivery" here.

Lorenz Jakober is a senior product marketing manager at Akamai
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20