Will IE10 have a Monopoly on Windows RT (ARM processor)?
Mozilla’s general counsel writes about the limitations browsers are facing with the new Windows 8 RT and how IE is the only browser that can run on native mode in this platform.

IE9 – Not Just For Work
We previously reported on Chrome usage spiking during the weekend in our Browser Wars updates. Now, StatCounter, is reporting that IE9 usage spikes on the weekend as well, showing that IE is making a comeback as a “play-time” browser with their latest release.

That Widget Might Cost your Webpage 1.34 Seconds
Think twice before adding a widget to you page. The author of this blog article discovers that the Facebook widget on the page is adding 84KB of data and 1.34 seconds to the load time.

Hydra: Big data at AddThis
AddThis handles 3 billion visits a day, resulting in 10 terabytes of data. Hydra is their real time analytics solutions which relies on custom-built distributed file system to store the data

F5 Load Balancer Now Supports SPDY
F5 announced first-to-market Mobile Application Optimization with SPDY support at Interop 2012. The goal is to improve mobile browsing experience by delivering content more efficiently and improving webpage rendering. This is an update to their Application Delivery Control, which now supports SPDY for browsers/clients that have incorporated the protocol.

Facebook Now Offers Cloud Storage
Right before the IPO, Facebook announces that they are now dabbling with cloud storage with the release of a new feature that allows groups of users to share a file. This puts Facebook in competition with other storage services such as Apple and Dropbox.

From China: Teleportation for Data Security
Physicists in China have been able to teleport entangled protons the longest distance yet – 97kilometers. While this is cool for whatever Sci-Fi fantasy you may have, the current focus is on the possibilities of satellite-based quantum cryptography, which could provide the most secure connections yet.

 Crate and Barrel joins The Home Depot and Le Monde in the elite group of companies that is actively working on improving web performance, to provide better end user experience and ultimately, improve their bottom line.

Back in 2011, we compared the website performance of the top Internet Retailer websites during the shopping season. Crate and Barrel did not do well in the study but instead of being angry at us, they called me and we had an amazing discussion about what went wrong.

This past week, I received another call from the Crate and Barrel team. “Hey Mehdi, we have been working on our performance and wanted to get your feedback!” I was shocked by the changes they had implemented. As I was charting their data for the last 6 months the first word that came to mind was “Beautiful”.

After the close of the holiday season, the Business and IT team at Crate and Barrel decided to make web performance a key priority for 2012 and I believe they are going to do very well this year.

Without any further interruptions here are the results:

• The time to Document Complete (or onload) improved by more than 49%
• The Total Page Load time improved by more than 39%

Here are performance measurements before and after the “Perflift” from 10 US nodes since Nov 2011.

Crate & Barrel Historical Web Performance Data

I picked a few dates to highlight the improvements:

Document Complete and Web Page Response

From a WPO (Web Performance Optimization) perspective the engineering team did a fantastic job at:

• Combining images with CSS sprites
• Reducing the number of third party tags, and loading the tags after document complete.
• Adding proper expiration headers to assets
• Relying on “Text” in place of images when possible
• Minification of JS and CSS files

Crate and Barrel Web Performance Optimization

The total downloaded bytes was cut in half:

Crate and Barrel WPO Downloaded Bytes

Crate and Barrel WPO Content Changes

I hear that the team is not done yet! I look forward to Crate and Barrel ranking very high in 2012 Shopping Season.

Congratulations to the entire Crate and Barrel Team for pulling off this “Perflift” (think Facelift, a rejuvenating SPA for website) successfully! Companies from around the world should take a close look and learn from Crate and Barrel, Le Monde and The Home Depot.

Mehdi – Catchpoint

Note: Crate and Barrel is a not a Catchpoint client.

Browser Wars Continue
Conflicting reports are out by two analytic companies regarding the IE rebound driven by the new release of IE9. Net Applications, which accounted for geographical usage patterns, daily usage, pre-rendered pages, and its own analytics into the analysis, reports that IE usage grew from 53.8% in March to 54.1% in April. StatCounter, using a different methodology in which they only count raw page views, reports that IE is still continuing to slide while Chrome keeps on gaining momentum.

Connection Speed Dropped in 8 of 10 Top Countries
Akamai’s most recent State of the Internet report shows that the global connection speed at the end of 2011 had dropped 14% from the previous quarter to 2.3 Mbps.  The US (ranked no. 13) connection speed fell 5.3% to 5.87 Mbps.

Optimizing Images for Mobile and Web
Jeremy Keith shares his tricks for optimizing the dConstruct conference website – including serving monochromatic images and focusing only on faces on pictures.

Push for Cloud Means More APM
As more companies are adopting the cloud as an infrastructure solution, there is a push in the Application Performance Management (APM) market to evolve to offer the visibility into the cloud that companies now have a need for.  Traditional and in-house APM solutions may no longer be enough to provide the insight needed to take control.

Relying on a Browser Hack? Make sure you test well and keep on testing.
One trick web developers use to pre-cache resources is to pre-fetch JavaScript and CSS files by relying on the <img> tag.  Eric Law of Microsoft clearly describes why this “hack” will not work and results in “request aborts”, resulting in slower webpage loads. The solution will not work on IE6 to 10 and Firefox 12.

Colocation Check List
Choosing a colocation site?  Here are a couple ‘out of the box’ questions to have on hand during the decision process.

Facebook Scuba project 
Using tricks from news feed implementation (delivering results of complex queries fast), engineers at Facebook built a real time internal infrastructure monitoring system capable of fast ad hoc analysis – delivering results that scanned hundreds of millions of samples in less than 1 second.

Who knew? JPEG uses less energy on your phone than GIF or PNG
Stanford researchers at the World Wide Web 2012 Conference presented data showing how free applications and top websites, even ones optimized for mobile, kill your smartphone’s battery. Energy consumption can be blamed mostly on advertising and browser downloading and parsing of Javascript and CSS.

Understanding Blekko’s NoSQL Database
The CTO of Blekko, a new spam-free search engine, describes the requirements it takes to build a search engine and datastore, including hardware and software options.  Using a new construct called Combinators, Blekko is able to manipulate data in the single cell level, instead of locking an entire row in as traditional NoSQL database.

Optimizing with the timeline panel
Want to troubleshoot site speed with the Chrome?  Check out this guide that shows how to use Timeline to find wasteful rendering operations.

The need for “Off the Shelf” data centers
White paper on the need for data center consumerisation and its benefits to business, quality. and cost savings.

Federal government agencies falling short on the IPv6 initiative
More than 99% of the federal government agencies do not support IPv6 on their public web infrastructure, falling short of meeting a 2010 directive from the the Obama administration. The key reason is lack of support for IPv6 by government contractors, including CDNs, network devices, and ISPs.

What is the right page speed target?
Level 3 blog post looks at why 2 seconds is the new page speed target and what is the correlation between page load speed and conversions. The article recommends sites to look at the maximum load time for a specific percentile, and focus on delivering pages in 2 seconds for specific user segments.

How Wayfair handled traffic load from a news article.
Wayfair engineering team discusses how they handled the unexpected traffic load from a news article which had a larger impact than Cyber Monday.

On other technology news, 10% of single men prefer getting the new iPad than a new partner.

A month ago, we wrote about how the Home Depot improved their web performance through a major “Perflift”. Today, we want to congratulate “l’equipe” at Le Monde for improving their website.

While The New York Times is the number one newspaper in US, Le Monde is the number one news source in France and for French speakers around the world. I have a deep and personal relationship with Le Monde; as it allowed me to stay informed while travelling around the world before settling in the US. I continue to be an online subscriber not only to have a deeper world news coverage but also to practice my French.

This year Le Monde’s IT team embarked on a major project to make the website faster. The key business driver was to increase web traffic by providing the best user experience. The development team rebuilt several parts of their system from scratch and optimized the front end code to achieve their performance goals. As part of the initiative, they modified how third party tags are rendered on the page in order to minimize their performance exposure and risks. This “Perflift” has a big payoff for the end users and the company as the new website blows away the old site in performance:

• The time to Document Complete (or onload) improved by more than 50%
• The Total Page Load time improved by more than 33%

Here are performance measurements before and after the “Perflift” from 5 European nodes (Paris, London, Amsterdam, Frankfurt, Milan)

Le Monde Homepage Performance

Web Performance Metrics Before & After

From a WPO (Web Performance Optimization) perspective the Le Monde engineering team did a fantastic job at:

• Reducing the number of CSS files
• Serving static content from cookie-less domains,
• Relying on CSS Sprites
• Domain sharding
• Loading the ads post document complete.
• Most importantly, loading JS asynchronously. They went from 700Kb to 100Kb of sync/blocking JS on the Home Page!
• Loading images asynchronously, triggering image load on user scrolling.

Le Monde Web Performance - Content Before & After

We hear that they have a few more tricks up their sleeves they plan to implement in order to keep improving the site speed.

Congratulations to the entire Le Monde Team for pulling off this “Perflift” successfully! Companies from around the world should take a close look and learn from Le Monde and Home Depot.

Bravo! Cocorico!

Mehdi – Catchpoint

Just when you thought your job never ended…
Engineers of Interxion, a collocation services company in charge of managing the Olympics, will work, eat, breathe, and sleep at their datacenter during the games.  The datacenter is equipped with sleeping pods, assuring they will always be onsite in case of a failure.

New strides for Quantum Computing 
Physicists from Max Planck Institute of Quantum Optics in Germany have been able to transmit qubits, a quantum bit, across 21 meter optic fiber network by exciting atoms to release photons.

ICANN misses its own deadline 
Because of internal software issues, ICANN went “temporarily” offline for four days.  This unavailability lined up exactly with the deadline for TLD applications and thus ICANN is extending the deadline one week to April 20.

HP PC sales grow faster than Mac in Q1
Q1 Reports show Hewlett-Packard domestic computer sales passed Apple at 4.6 million PCs to 1.7 million Macs with a growth rate of 6.6% and 5%, respectively, over the previous quarter.

New Concurrent User Record for Skype
On Tuesday, April 10, Skype reached a new high record of 40 million concurrent users on their service.

Cloud Perflift for 2012 – Will have Cost Impact
Traditionally, cloud computing platforms offered either a choice between old technology, low-cost public storage and more expensive, higher performance private clouds. Now, as buyers are wising up to performance impacts on revenue, there is a push in the market for solutions that offer a high availability, secure alternative for the smaller consumer. This hybrid cloud demand will drive up the cost of low-end solutions up.

Network Latency is No Longer an Enterprise Concern
As more and more people rely on richer web applications for entertainment and communication, network latency has a bigger impact on the home user. However for the typical user, unlike someone on trading networks on Wall Street, one cannot simply ask their provider for a better connection. Luckily, new products are development that help manage the consumer’s network – such as setting priority to certain packets that need to be delivered faster.

Amazon S3 Storage Growth for Q1 2012 Triggers Conflicting Press
Amazon reported growth in the first quarter of 2012 to their S3 services from 762 billion stored objects at the end of Q4 2011 to 905 billion objects. However, the data  provided on the blog was confusing as it provided yearly numbers prior to Q1 2012 – resulting conflicting news reactions. The question still remains did S3 growth slow down or accelerate?

Web Performance for Everyone!
Many network optimization techniques developed by major high scalability internet companies are now available as consumer products. Now, websites have an easier time scaling to more traffic without having to develop in house solutions.

In a DNS zone, every record carries its own time-to-live, so that it can be cached, yet still changed if necessary.

This information is originally served by authoritative servers for the related zone. The TTL is represented as an integer number of seconds.

At first sight, the mechanism looks straightforward: if the www.example.com record has a TTL of 30, it’s only valid up to 30 second, and caches must fetch it again if requested after this delay.

Chained caches

DNS caches can be chained: instead of directly querying authoritative servers, a cache can forward queries to a another server, and cache the result.

Having 3 or 4 chained caches is actually very common. Web browsers, operating systems and routers can cache and forward DNS a query. Eventually, this query will be sent to an upstream cache, like the ISP cache or a third-party service like OpenDNS. And these caches can also actually hide multiple chained caches.

In order to respect the original TTL, caches are modifying records as they forward them to clients. The response to a query that has been sitting in a cache for 10 second will be served with a TTL reduced by 10 second. That way, if the original TTL was 30 second, the whole chain is guaranteed to consider this record as expired as the same time: the original meaning of the TTL is retained no matter how many resolvers there are in the way.

Well, not exactly. A TTL is just a time interval, not an absolute date. Unlike a HTTP response, a DNS response doesn’t contain any timestamp. Thus, requests processing and network latency are causing caches to keep a record longer than they actually should in order to respect the initial TTL.

A TTL being an integer value makes things even worse: a chain of N caches can introduce a N second bias.

In practice, this is rarely an issue: TTLs as served by authoritative servers are considered indicative, and not as something to depend on when accurate timing is required.

The decay of a TTL, as seen by different pieces of software

How does the TTL of a record served by a DNS cache decays over time?

Surprisingly, different implementations exhibit different behaviors.

For a record initially served with a TTL equal to N by authoritative servers:

• Google DNS serves it with a TTL in the interval [0, N-1]
• dnscache is serving it with a TTL in the interval [0, N]
• Unbound serves it with a TTL in the interval [0, N]
• Bind serves it with a TTL in the interval [1, N]
• PowerDNS Recursor always serves it with a TTL of N

Hold on… Does it mean that how frequently an entry will actually be refreshed depends on what software resolvers are running?

Sadly, yes. Given a record with a TTL equal to N:

• Bind and PowerDNS Recursor refresh it every N second if necessary
• Unbound and dnscache are only refreshing it every N + 1 seconds at best.

TTL Zero

Although a TTL of zero can cause interoperability issues, most DNS caches are considering records with a TTL of zero as records that should not be cached.

This perfectly makes sense when the TTL of zero is the original TTL, as served by authoritative servers.

However, when a cache artificially changes the TTL to zero, it changes a record that had been designed for being cached to an uncachable record that contaminates the rest of the chain.

dnscache and a 1 second record

To illustrate this, a Linux box has been setup with a local DNS cache. Unbound has been chosen, but the last component of the chain actually makes little difference. Even web browsers caches have a very similar behavior.

Queries are forwarded to an upstream cache on the same LAN, running dnscache, and outgoing queries are recorded with ngrep. The same query, whose response has a TTL of 1, is made at a 10 queries per second rate.

The X axis represents outgoing queries, whereas the Y axis is the time elapsed since the previous query.

Even though a TTL of 1 is served by authoritative servers for this record, a large amount of responses are cached and served for 2 second. This is due to the local Unbound resolver.

However, there is also quite a lot of responses that haven’t been cached at all. It happens when dnscache serves a response with a TTL of 0. Since it happens one third of the time, this is suboptimal and not on par with the intent of the authoritate record, which is served with non-zero TTL.

And although dnscache and Unbound are handling TTLs the same way, we can’t expect their caches to be perfectly synchronized. When the local cache considers a record as expired and issues an outgoing query, the upstream server can consider it as not expired yet, just in the middle of the last second. What we get is a constant race between caches, causing jitter and outgoing queries sent after 1 second.

How different implementations can affect the number of outgoing queries

The following experiment has been made by observing outgoing queries when sending 10,000 queries for a record with a TTL of 2, at an average 10 qps rate, to a local cache, using Google, OpenDNS and Level 3 as upstream resolvers.

Here is the number of outgoing queries that were required to complete the 10,000 local queries using different services:

When using Google DNS (blue dots), queries are effectively never cached more than 2 second, even locally. This is due to the max TTL returned by Google DNS being the initial TTL minus one.

When Google DNS returns a TTL of zero, we observe the same jitter and the same slew of queries that couldn’t be served from the local cache.

OpenDNS (orange dots) has the same behavior as dnscache and Unbound, with TTLs in the [0, N] interval. Our initial TTL with a value of 2 actually causes 3, 2, and 1 second delays between request, plus a sensitive amount of consecutive outgoing queries due a null TTL.

Note: Since the original post OpenDNS resolvers have been updated to behave like Bind: TTLs are in the [1, N] interval.

Level 3 (green dots) is running Bind, which returns a TTL in the [1, N] interval. Bind caches records one second less than dnscache and Unbound. Because our local resolver is Unbound, queries received with a TTL of N are actually cached N + 1 second. But as expected, the frequency of required outgoing queries very rarely exceeds 3 second.

What the correct behavior is, is out of scope of this article. All major implementations are probably correct.

But from a user perspective, with only 2 caches in the chain, and for a given record, the same set of queries can require up to 3.5 more outgoing queries to get resolved, depending on what software the remote cache is running.

With CDNs and popular web sites having records with a very low TTL (Facebook has a 30 second TTL, Skyrock has a 10 second TTL), the way a cache handles TTLs can have a sensible impact on performance. That said, some resolvers can be configured to pre-fetch records before they expire, effectively mitigating this problem.

The OSX cache

MacOS X provides a system-wide name cache, which is enabled by default.

Its behavior is quite surprising, though. It seems to enforce a minimum TTL of 12.5 seconds, while still requiring some outgoing queries delayed by the initial TTL value, and some consecutive ones. Resolving the 10,000 queries from the previous test took only an average of 232 outgoing queries.

Using Google DNS, OpenDNS and Level 3 as a remote resolver produce the same result, with the exception on Level 3 (Bind) avoiding frequent (less than 1 sec) consecutive queries during the time other return a TTL of zero.

Cache Insights From Steve Souders
Maximizing cache utilization and compressing text responses are key web performance technique. However, not all browsers cache compressed files the same way. IE and Safari store items in cache uncompressed, while other browser store the compressed file – maximizing the utilization of cache size limits.

The New 308 Permanent Redirect HTTP Status Code
Eric Law from Microsoft talks about the challenges of the newest HTTP status code, the 308 Permanent Redirect.

AOL Retires 9,500 Servers 
As part of the Server Roundup Contest to minimize energy utilization, AOL decommissioned 9,500 servers in less than 6 months to save $5.05 million.

Impact of 3rd Parties on US Site for Chinese Users
Everyone in the web performance community should know by now that 3rd party tags that load synchronously will impact end user experience. Yet, the majority of top US sites still have not solved this problem. Of course users from China are impacted the most due to the “Great Firewall“.

At Dyn, we obsess about network performance and the proof is in the tools we have built over the years to constantly monitor how our network is running. With a globally deployed Anycast DNS network, constant internal and external monitoring of our performance is critical. This constant surveillance of our network ensures that we keep providing top-notch services. One of the biggest challenges is being at the mercy of third party providers and their monitoring platforms when monitoring our Anycast network.

After all, we cannot monitor ourselves and be fully objective with the data.

One of the issues with using external monitoring providers has always been the ability to pull data from the monitoring provider within an acceptable interval. We really want to be seeing data within a few seconds of a test run completing so that we can correct any issue just as soon as it happens. For a long time, we’ve been working with monitoring providers that can get us our data in five or ten minutes at minimum and on a dashboard they render – hardly usable for our operations.

As we’ve mentioned before, one of our favorite monitoring providers is Catchpoint and thanks to their Data Push API, we’re able to receive a constant stream of feedback from their 50 global monitoring nodes in real time. Every five minutes, a Catchpoint node performs a series of tests against our DNS servers and instantaneously relays that information to the central Catchpoint collector, but also ships a copy of the results to a webserver on our network so we can begin reviewing those results immediately.

Enter the challenging part: how do we build a dashboard with ACTIONABLE data with 50 data sources and 4 targets (over 200 data points) over an hour’s time? Enter Alex Sergeyev from our Dyn Labs team, some ZeroMQ love and work with Websockets and D3.js. Alex built us a very slick visualization application that allows us to really see what’s happening in real time.

The design of the perfect dashboard

We decided to create a system to deliver the stream of data from Catchpoint API down to all connected web-browsers in forms of 50+ little charts. In the past, we have been shipping the data to a Graphite server as well, so we wanted to keep that running too.

We built three parts of the system to scale them independently as needed:

• Catchpoint API receiver that analyses data and sends simple parsed lines to ZeroMQ PUB socket.
• Websockets server that receives data from a ZeroMQ SUB socket and also keeps keeps state of last probes per monitoring node in a memory-mapped file (for some flavor of persistence across restarts).
• A Graphite “shovel” which also receives data from the same ZeroMQ SUB socket and sends results to a Graphite server.

We used Perl for both of the web services parts, because it gives us a way to code simple applications very fast. Both web services were implemented as simple AnyEvent servers.The design above allows us to have sub 100ms latency between receiving data from monitoring nodes and sending it out to a connected web sockets.

After building out the backend, most of the development time was spent to create visual representation of the data we got. We wanted to be flexible with our view and decided to use D3.js to help us with that. It’s an amazing tool which allows to make truly data-driven visualizations. We started from simple line/area charts to emulate visualizations that we are used to look at in Graphite and RRD based tools. We even wrote a SmokePing emulation library for D3.js, used it for our dashboard and plan to release it via our GitHub page in the future.

Here are few examples of charts that it generates (dashboard size):

Smokeping Dashboard WidgetD3.js

Taking these basic charts and the smarts of our UX genius Lara Swanson, we transformed boring sets of 50 charts on a page to a display that ended up looking neat and pretty on either laptop screen or Internet TVs around the office. We decided to group charts by monitoring region and sort each section (credits to D3.js for making it simple) by mean response time, plus twice of standard deviation to “promote” trouble spots to the front of each grouping dynamically to make the data actionable!

Our current dashboard is not quite final but we’re proud of what we quickly built:

• a scalable sub 100ms data publishing system that kept our existing Graphite integration working and added our new dashboards.
• a simple, dynamic view of all available data (over twenty thousand data points) aggregated to help us know if there are any regional or local issues affecting the quality of our DNS services.
• an easy spot for our capacity planning and P&O engineers to look at when they are looking to make improvements to the network.

This post originally appeared on the blog for Dyn, the Internet IaaS Infrastructure-as-a-Service leader that features a full suite of DNS and Email Delivery solutions. Follow on Twitter and like on Facebook.

How OMGPOP Scaled its Infrastructure
The creators of Draw Something were able to scale their infrastructure to support 36 million users in 3 weeks by using a NoSQL database and cloud computing.  Engineers were able to deploy and reconfigure databases live with no hiccups.

Microsoft Proposal on HTTP 2.0 – Focus on webpages and applications
Microsoft stirs the browser wars by releasing their proposal on HTTP 2.0 which is based on SPDY, but goes further to support not only webpages on browsers but also applications. In their opinion SPDY was a good start, but not good enough of a solution.

What is Impacting Inter-cloud Communications? The Internet or Business Practices 
A recent survey by Nasuni on the data transfer between the top 3 cloud providers (Amazon S3, Microsoft Azure, Rackspace) shows high latency in communications.  Transfer time ranged from 115 hours (Amazon S3 to Rackspace) to as little as 4 hours (Azure to Amazon S3) for 12 terabytes of data. One wonders what is causing the slowness in getting the data out of Amazon?

LinkedIn’s Databus: A Low Latency Change Data Capture System
Siddharth Anand from LinkedIn covers the need for Change Data Capture Systems and how LinkedIn’s own Databus system works. Databus distributes in near real time the updates from a primary data store (like a relational database) to a number of secondary data stores.

February came and went so fast we forgot to give you a quick update on what we have been up to!

Empowering Data Analysis

Our in house NoSQL/In-Memory reporting platform allows customers to run statistical analysis of the data in real time and it stores historical data for 2 years+ since the first day of the launch. However,  it had one limitation – customers could not chart and report at the 5 minute breakdown past the last 40 days. Well no more!

Last December, our engineering team embarked on a big project to enhance our backend systems and unleash the stored data to Catchpoint end users. Our data warehouse already stored the data we needed, however, we had to drastically change how data was stored and rewrite the report service to properly access and process it.

Starting this week Catchpoint customers are able to chart up to the 5 minute interval and even chart a scatter plot at the minute level on historical data going back 2 years. Yes, they can even chart a scatter plot from the data collected in 2010!

Mobile Playback:

Customers can now test the mobile sites and infrastructure using a mobile emulator that monitors requests triggered on a browser. We support iPhone4s and Android, and have the ability to playback a mobile HAR file.

By the time we are writing this we are already 2 weeks from the March release with more cool new things coming!

The Catchpoint Team.

Cloud Computing Saboteurs and Counter Intelligence
Everyone is leasing datacenter space these days, including major internet giants like Google. Fear of neighboring competition sneaking a peak at top secret hardware caused Google to turn off the lights and close the blinds in their part of the corridor. Only Google employees can ascend into the darkness, equipped with a head lamp of course.  We wonder though, if a competitor made past security and into the Google datacenter – what would stop them from bringing a flashlight?

This fear is not without reason however, as everyone is trying to build a better, bigger, and more efficient cloud computing solution. A recent study released by Huan Liu at Accenture Technology Labs, predicted that Amazon’s Web Services is powered by at least 454,400 servers across seven global data center hubs.

Google Datacenter Going Green
Google is not only saving electricity by cutting the lights – they are also taking a creative approach to going green by spraying recycled waste water on servers to keep a new datacenter cool.

Stormy Days
Leap Day 2012 was the trip and stumble felt around the world for Microsoft Azure.  A calendar bug caused failures in Azure clusters all over the world when all regions entered Leap Day at the same moment (proving redundancy is not perfect). Combined with the recent Amazon EC2 hiccups, companies are now re-considering their cloud strategies and looking at a multi-vendor approach.

If you don’t like it, build it yourself
The Chromium team ran several performance tests for DNS resolution on IPv6 enabled machines and it was not happy with the resolution speed of the OS (Windows, Mac, or Linux).  As a result the team is working on building their own DNS stub resolver to distribute with Chrome browser.

Higher Resolution, Smaller Screens, More Performance Headaches
The release of the new iPad, boasting an even-more-retinal-display, is causing a surge in the population of people surfing the web with HD screens. This will have huge impact on how images are served as normal resolution images appear fuzzy on HD screens. The solution isn’t as clear-cut as just increasing resolution – images account for about 60% of total downloaded bytes and tripling the resolution to keep up with iPads will cause the page size to double!

Windows 8’s Complete Makeover Extends to IE
Microsoft engineers not only rethought the desktop experience in building Windows 8, but they also did a complete makeover in IE, optimizing the browsing experience for keyboard, mouse, and touch. Look for a completely new way to integrate websites and apps as designers sought to make heavy workflow as seamless as possible.

Cloud Pricing Wars – Game On
Last week,  Amazon AWS reduced their prices by as much as 42% for certain reserved EC2 instances. This was the 19th time Amazon lowered their prices and it triggered a pricing war in the cloud space. Google was quick to announce on the same day their new reduced prices that are slightly lower than AWS. While 3 days later Microsoft followed up with prices cuts on Azure.  While the lower prices will help with the migration to the cloud, companies shopping for a cloud provider are still challenged by the difference in the computing instances packages which make any comparison quite hard. Maybe the Cloud Providers need to introduce a price/unit concept like the products in your local grocery shelves.
Web boom causes demand for new data centers
A new report out by Campos Research and Analysis on behalf of Digital Realty Trust shows a growing demand for more data centers worldwide. In North America 92% of the respondents claimed they will probably to definitely increase their colo spending this year in the US, Europe, Asia Pacific, and South America, in order of planned expansion.  85% of respondents worldwide intend to increase resources as opposed to 82% at the end of 2010.
Creative Crash Pages Making the Best Out of a Bad Situation 
Despite ample prevention, all sites are bound to crash or be unavailable at some point. How a company handles such scenario, can make a difference.  Bonobos showed during the Small Business Summit 2012 that with a little bit of humor and a dash of humility, a well thought out Down Page can turn a revenue crisis into a marketing opportunity, even if the crash happened on Cyber Monday.  Other retailers should take notice and plan for similar action in 2012 – it pays off to be ready.
IBM Processor Can Move Data at 1 Tb/s 
As computing technology is advancing and servers pack more and more processors, the network overhead or latency is becoming a growing bottleneck. IBM released a new solution, the Holey Optochip, which can move data at 1 trillion bits per second (1 terabyte per second) by firing lasers through the 48 holes in the chip. The Holey Octochip (not for sale) is 8 times faster than the fastest chips in the market today and 10k times faster than a standard 100 Mb/s Ethernet connection.