ITC Infotech acquires Blazeclan Technologies to enhance Multi-Cloud services and fast-track digital transformation

4 Key Takeaways from Andy’s Keynote at AWS Summit 2014 – Why is World Moving to Cloud so Fast?

Snapshot: Andy Jassy Key Note San Francisco 2014

So What Made the AWS San Francisco Key Note by Andy Jassy So Different and Value Generating? Well it’s something which AWS is renowned for, on how they do business. Looking at every customer need and creating the simplest of Abstractions to solve the most complex Customer Problems. In part 1 of the Key note Andy Jassy Deep Dives with several  examples into those 4 Important reasons as to why people are moving to the Cloud so Fast. In the 2nd  part he moves on to crystallizing those 7 Key ways on How Enterprises are using Cloud, what are those pain points which enterprises are fearful of, and how exactly AWS Caters to solve them. See what Silicon Angle Guys are talking about this AWS Re: Invent


The Key Note Started with Some very Insightful Gartner Quotes on AWS & the Cloud:

“AWS has 5 Times more the compute capacity in use than the aggregate total of the other 14 Providers”  (The other 14 providers include the players listed in Gartner’s Magic Quadrant Calculation)

“[Enterprise customers are] skipping the years of early getting-their-feet-wet, and immediately jumping in with more significant projects, with more ambitious goals…”

“Increasingly organizations are asking not what can’t go to the cloud, rather than what can?”

“As 2014 Dawns, we’re moving into an era of truly mainstream adoption of cloud”


So Why is the World Moving to the Cloud so Fast? AWS Senior VP Shares his Thoughts:


1] The AGILE Creator:

When we talk about StartUps, what’s that main brownie point that truly triumphs over Larger Enterprises? It’s the Pace of Innovation and the ability to implement that idea. But for enterprise it takes around 6-25 weeks to move into new projects, this is too long a time and ideating minds fearful of this time drag are resenting from thinking outside the box.

So what’s that other major issue for New Experiments that Enterprises Fear so much?

The cost to Create (Large amounts of Compute, Storage, Database) environments to run these experiments are a lot, so basically infrastructural costs. More devastating is when your experiments fail(and they will fail half the times if you’re pushing the envelope) & you suffer the collateral damage of investing in so much of Software & hardware.

But with cloud computing Enterprise IT teams can now focus their Horse power in driving the innovation, creating new ways to improve Customer Satisfaction. More over their time lines to see actionable results will shorten as the cloud takes care of all the initial heavy lifting. Also IT Teams can now focus much less on racking and stacking software and hardware.


2] A Vast Platform Breadth to Work with:

With Cloud, Especially on AWS You have a very broad functionality on Cloud. With more than 10 Regions spread across the world with more than 50 Points of Presence, it provides the perfect infrastructural base for all the Services on top of it. In the most fundamental levels of infrastructural service, lie the building Blocks of Compute, Storage & Database. Following this you have the networking capabilities of Virtual Private Cloud for Your VPN, Direct Connect & Route 53. Above this you have the Analytics & Application level Tier with services like EMR & Kinesis. Above this lies the management and application level capping with the Containers & Deployment services like CloudWatch, Elastic BeanStalk, Cloud Formation, Opsworks & more. For a Comprehensive Tier Level understanding of AWS infrastructure, check the Diagram.

Complementing these several services are the +25 product categories with over 1400 product listings, grown by over 90%. These SI & ISV partners have their software & hardware offerings on AWS Marketplace pre-configured to work seamlessly with AWS EC2(Compute) and ready to be deployed for the ease of AWS users.

But What’s the point in deploying Services if they are not feature rich?

Well AWS Cloud proudly answers this question, with more than 280 feature releases in 2013 & already 82 feature releases in 2014. If you look at specific services Redshift, Amazon’s Peta-Byte Data warehouse for example, has released more than 56 new features since its launch in Feb.2013. AWS Instance Family Evolution: Building instance families as customer requirements pour in. The 5 Types of Instances can be categorized as:

  • General Purpose instance type (M1)
  • Compute Optimized (C1 & CC2)
  • Storage & IO Optimized(H1 & HS1)
  • GPU Enabled(CG1) [For gaming & computer graphics]
  • Memory optimized ( M2 & CR1)

[11 Key Criterias for Cloud Migration]


However let’s look at some of the New Generation Changes over the last 6-9 Months:

  • The new M3 instance type has SSD based storage, the latest processors with lower price than M1.
  • The C3 instance type with enhanced networking capabilities can deliver up to 4 times better packets per sec than c1 instance, with hardly any jitter and an overall 60% better performance than C1. It even has high memory for High performance application requirements, another feature request addressed.
  • The I2 instance upgrade from HI1: With 56% more SSD than HI1 with IOPs increased capabilities, the largest I2 instance type can process up to 3,65,000 4kb random read IOPs & 3,15,000 4kb random write IOPs.
  • The new G2 family has more instance sizes than old CG1 at lower prices and improved performance.
  • New instance Type R3 for Memory optimized workloads with the following features to look at:

– SSD based Storage
– 8:1 ratio of memory  to CPU
– 244GBs of RAM
– Upgraded networking capabilities like the C3 instance type

To facilitate these Services you need a World Class Network, with AWS You get:
  • Higher packets-per-second
  • Low Jitters
  • EBS optimized instances
  • Virtual Network interfaces
  • High Throughput, that is inter instance lower latency
  • Physical Placement optimization
Along with this a Global Footprint Enhances the Networking Capabilities:
  • 10 regions
  • 26 availability zones
  • 53 edge locations
So How are Organizations Leveraging this Huge Platform?

> SUPERCELL:  They track their gaming analytics and measure what users are doing using AWS Kinesis(Amazon’s Big Data in real time).

> NASA: When the Mars ROVER was landing on the planet, NASA realized that even the public would want to watch it as it happens. So they used the basic S3 & EC2 with pre-configured Adobe Flash media servers from AWS Marketplace, used ELB to seamlessly manage the load as more users joined the live stream with CloudFotmation ready made stacks to deploy as loads increased.

>NITENDO:  To manage their Social Networking Application Miverse, embedded inside these games, they required low latency catered by Route 53 & analyzed the data using AWS Redshift. For high availability they deployed along multiple Regions & Availability Zones within VPC environments connected using VPNs using Elastic Network Interfaces.


3] Continual Iteration & Innovation on the Cloud:

AWS Cloud Continuously upgrades infrastructure some of which gets automatically updated to your configurations without you having to move a muscle! We spoke about how important it is to launch feature rich services a little earlier, but let’s look them in detail:

Snapshot from Key Note Speech, AWS Innovates


Let’s Delve Deeper into some of these Customer Requirements answered in Features:

Dynamo DB, the NOSQL Service has had a series of releases some of them are:

  • global & local secondary indexes, to improve query flexibility
  • item level access control
  • parallel scans
  • batch rights and many more…


 [A New Approach to Cloud Migration – 4 Key points to Consider]


AWS Also has launched whole New Services based on Customer Needs:

1)Amazon AppStream for helping companies optimize streaming of applications or their components (containing high graphic content) to their end users based on their device types. Appstream uses its own protocol to know what the network connection & device types are on the fly. There are many customers successfully using appstream, like eve online, a gaming company which uses this service to allow users to create high graphic intense avatars. Trugolf, the worlds leading swing simulator uses appstream to cut down on expenses of their swing simulator by using a sensor combined with appstream to reach bigger audience at a lower cost.

2) Amazon WorkSpaces: Desktop Virtualization problems using VDI(centralized managed desktops) solutions were many like: maintaining hardware, software, date centers & other infrastructure pieces. It was also an expensive bargain and to top it off had performance issues. But using Amazon WorkSpaces all these issues are automatically taken care of, at half the price of VDI solutions. The success of work spaces can be seen as over 10,000 customers have signed up since November 2013. Andy Jassy also announced that they are making Work Spaces public.


4] So We keep talking about How Cost Effective AWS Cloud is, Let’s Check this Out

So when AWS says it passes it’s cost benefits to its customers, what does it mean, well a number of advantages, so of which you know, so of which you might not:

a) Pay only for what you use! Don’t lay off that capital upfront for data centers, with cloud you can turn cap ex to op ex.

b) Economies to scale, passing over the value to the customers, providing costs that amount to be cheaper than in house costs. Check out the latest 42nd Price reduction effective as of 1st April 2014: (more on price cuts)

  • 36%-65% on S3 depending on tiers worldwide.
  • EC2 reductions to 30-40%.
  • RDS reduction 28% average.
  • Elasticache reduction of 34%.
  • EMR  reduction 27-61%( New instance types (support available) to run with EMR)

c) A variety of Pricing models to fit every requirement. Pricing model choices which support variable as well as stable workloads (You can consume AWS by the hour like EC2) using on demand pricing. Many Companies nowadays are combining the different pricing models. Example: For steady state work loads companies make a small upfront commitment to gain a 50-75% reduction than what you get on demand, using reserved instance pricing and for opportunistic workloads which can afford minor interruptions you have spot instances which you can bid for. Customers run reserved instances for their steady state work loads, as they end up saving so much on a per hr basis. They then fill in the rest of their work loads by on demand or spot. AWS also has various tier pricing options, volume discounts & even custom discounts for larger players.

We spoke so much about cost reductions, so let’s end part one of the Keynote discussion with a few customer views on how they benefited from these price reductions:

“Without AWS we would have needed to raise $7.1M to launch instead of $1.8M” – Mortar

“Migrating Data Centers to AWS Contributes to a global savings of 100 Mil dollars in infrastructure costs”  – Dow Jones

“The move could save the navy as much as 60% versus the cost of hosting and managing data in its own data centers” – US Navy

Stay Tuned For Part II of this interesting keynote where Andy takes on How Enterprises are thinking about using Cloud in 2014.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.