Digital Walkabout

All good things come to those who wait ... in cloud

Corporate IT execs are locked in debate over whether their companies can really save money by moving some applications to the cloud. The IT community is riddled with stories of companies that launched a public cloud app, only to build a private cloud once costs flew out of control.

So does cloud computing really save money? And how do you choose the cloud platform best for your needs?

When Brickfish, a Chicago-based social media marketing software provider, moved its headquarters from San Diego to Chicago in 2009, the company switched from an on-premises data center to Rackspace Hosting's managed service model. But it found the cost savings to be negligible.

Renting servers with Rackspace still cost the company about $47,000 to $52,000 a month, said Michael Mullarkey, CEO of Brickfish. When it was time to renew its Rackspace contract, Mullarkey thought he could cut expenditures and improve Brickfish's ability to innovate if he moved the company's entire technology stack to NaviSite's NaviCloud.

The shift also allowed Brickfish to partition off certain virtual machines for some of its larger clients that wanted their own virtual private clouds for security reasons. One such company was retailer Vera Bradley, which used Brickfish's social marketing app to create a page on Facebook that features its housewares products.

With the cloud migration to Navisite, Mullarkey saw significant cost reductions. Brickfish went from spending just under $700,000 hosting and maintaining boxes with Rackspace to under $200,000 with NaviCloud. Mullarkey was able to pump that $500,000 in savings back into Brickfish's own research and development.

Others do agree that there are cost savings from the cloud, although not necessarily immediate.

Bluebird Auto Rental Systems, a Dover, N.J.-based firm that handles credit card transactions for auto rental centers worldwide, needed a cloud provider with PCI DSS certification. Shortly after Amazon Web Services (AWS) became PCI DSS-certified, Phil Jones, vice president at Bluebird, said the company jumped from hosting physical servers with OpSource to AWS EC2 public cloud.

By simply eliminating the server-leasing fees it had with OpSource, Bluebird cut costs from 60% to 70%. However, the money saved from moving to EC2 didn't magically stream in.

"After we were in the cloud for about six months and were sure this was the way we wanted to go, we took advantage of [AWS EC2] reserved instances and now get a much lower rate," Jones said. "The break-even point is about six months. That's where we're saving money for the rest of the three-year term."

Cloud value isn't just about the almighty dollar The cost savings of moving to a public or private cloud is one thing, but you're getting much more value than cold, hard cash, said Mitch Greenwald, CEO of CloudBakers, a Chicago-based integrator. The trick is to choose the right cloud service model and cloud provider in order to get that value.

"SaaS offers an extremely high value proposition, and PaaS [offers] accessibility," he said. And while IaaS has its benefits for some companies, it won't necessarily reduce capital expenditures. "With Rackspace or AWS, you still need to manage servers, monitor your network and maintain it -- even though it's a cloud," Greenwald added.

Software as a Service brings big data analytics to the masses

When it comes to managing on-premises systems, virtualization and cloud computing shops want quick answers to complex queries -- without dealing with the underlying infrastructure.

Some of those shops have taken a leap of faith and adopted emerging products that offer such "big data" analytics features, delivered via Software as a Service (SaaS).

"They can simplify views of very complex information into a webpage that anyone can look at," said Nathan Smith, senior engineer at Centered Networks, a San Francisco-based Desktop as a Service provider.

Smith has experimented with CloudPhysics, a cloud-based beta offering that aggregates data using big data techniques. It aims to compare data from different customers to provide deeper insights into virtual and cloud infrastructures, and it draws from VMware Inc.'s vCenter statistics rather than machine logs.

"You don't need to be an expert … because it's broken down in [a] very simple, tabulated format," Smith said.

Centered Networks hopes to more easily troubleshoot storage performance issues in its VMware infrastructure and speed up its customer service responses with CloudPhysics' big data analytics Software as a Service once it hits general availability, Smith said.

"We can see from a storage perspective how the storage thinks it's doing as far as performance, but that doesn't necessarily tie up with ESX thinking that the storage is performing the way it needs to," he added.

How big data analytics Software as a Service works
Other companies offering big data analytics Software as a Service include Sumo Logic, Splunk, AppFirst and ScaleXtreme. The information their services gather is anything but simple.

LimeLight Networks, a Web presence management company, supports customers through its own SaaS offering, behind which is a pool of about 600 physical servers. Within that infrastructure is a Web content management business, which accounts for about half of those servers. The company became an early adopter of Sumo Logic's service after failed attempts to aggregate its machine logs on-premise.

"I can start typing in a query that will give me data across two different data centers, multiple different apps, and it's all essentially happening in real time," said Tom Cignarella, vice president of SaaS technical operations for Tempe, Ariz.-based LimeLight Networks, an early adopter of Sumo Logic's service.

LimeLight has so far deployed Sumo Logic on the servers in its Web content management business.

"We'd always had issues with disk space and how to ship things around and all the headaches that come along with that," Cignarella said.

Sumo Logic, officially launched in January, aggregates machine log data in the cloud and makes it available for search by customers. In the background, it uses some components of the Hadoop Distributed File System and Cassandra, along with its own distributed database, dubbed Bruno. Also, Sumo Logic uses a patent-pending algorithm it claims can shrink a million log messages down to 10 to 15 categories, thereby getting to the root cause of issues in the customer's cloud infrastructure faster.

"It's made us better at doing our job," Cignarella said. "We could always dig through individual log files, but this has made it easier to narrow things down more quickly."

Big data analytics Software as a Service still evolving
Some of Sumo Logic's capabilities remain a promise for now, as it's still a very young company. For example, Cignarella and his staff can run search queries on aggregated log data, but dashboards that would track the most common queries are still on Sumo's roadmap. Comparative analysis of LimeLight's log data with other Sumo customers, another facet of Sumo's big data analysis, is also something Cignarella hasn't looked into. Another common aspect of big data analytics, anomaly detection, has yet to be incorporated into the product as well.

Sumo Logic officials declined to comment on product features that haven't yet been announced. CloudPhysics also has a ways to go before it's a finished product, Smith said.

"It's kind of an evolving product, and it's a process for us, where we're trying to figure out what else we can get from it," he said.

CloudPhysics' software boils down its assessments of individual customers' environments and presents them in a tiled interface, with each tile (called a "card") showing information about a different aspect of the infrastructure. Centered Networks has been helping CloudPhysics develop its storage performance card, for example, and during that process, ideas for new cards have sprung up, Smith said.

"Right now the card we're working on is changing the HA configuration, so [like a] simulator … it shows you what the actual impact would be to your hosts," he said.

What the recent VDI acquisitions mean for you

Since April, there have been four major acquisitions in the desktop virtualization market. For many administrators, that means more choices and the reassurance that VDI offerings will continue to improve.

 

Dell, VMware and Citrix Systems made moves that will transform IT's options for thin clients, VDI software and infrastructure setup. Some of these acquisitions involved major vendors, and others featured smaller niche companies, but each changes the market when it comes to virtual desktop infrastructure (VDI) feature sets, client virtualization offerings and physical desktop management.

 

With all the wheeling and dealing going on between desktop virtualization vendors, it's time to look at how these acquisitions will affect IT pros managing VDI environments.

 

April: Dell acquires Wyse for thin clients 

It all started with Dell's acquisition of Wyse. The move gives Dell an edge in the VDI hardware market, experts said, particularly because Wyse thin clients and zero clients can integrate into any virtual desktop environment.

 

Acquisitions zero in on management

These acquisitions give a boost to virtual desktop management. Virtual Computer's NxTop provides Citrix with the endpoint device management features XenClient lacked. Dell could use Wyse Stratus as a device management platform. 

 

Last, VMware's acquisition of Wanova shows that many IT pros still need to manage physical desktops as part of their VDI environment.

 

With this deal, Dell finally has its first in-house enterprise thin client offering to get on par with Hewlett-Packard. It also adds some intrigue to the competition between Cisco and Dell, because Cisco's thin client offerings are based on Wyse technology. Plus, Dell's existing partnership with DevonIT could go by the wayside if Dell discontinues use of the company's OEM clients.

 

May: Citrix buys XenClient competitor

In a move that some industry observers said eliminates client hypervisor competition for Citrix, the company bought Virtual Computer. Citrix invested in Virtual Computer in 2009 but then built its own Type 1 client hypervisor, XenClient. Still, experts said the startup's NxTop outshines Citrix's own product.

 

This deal signals to VDI managers that client-side virtualization is a valuable alternative to thin clients. Citrix will integrate NxTop technologies into the new XenClient Enterprise Edition. The company also has the resources to move NxTop into more hardware manufacturers and devices -- perhaps even mobile.

 

May: VMware gets physical with Wanova

VMware bought Wanova for its Mirage single image management technology, which it will integrate into VMware View. Mirage keeps a centralized copy of the user's desktop from every device they use, which can then be loaded into View in disaster recovery situations. For mobile workers, this will be especially useful for recovering stolen or failed PCs on a tablet, for instance.

 

This acquisition is an interesting one for the desktop virtualization market because it indicates that VMware isn't ignoring physical desktops. That's because Mirage works with any endpoint -- physical or virtual.

 

July: Dell makes big splash with Quest acquisition

Dell made another significant acquisition toward the end of summer. With Wyse firmly in its pocket, the company bought Quest Software to boost its caché as more than just a hardware giant. VDI shops can now look to Dell for all their needs: PowerEdge servers, EqualLogic storage, PowerConnect networking, Wyse thin clients and host-side virtual desktop software.

 

Some IT pros assumed this deal meant Dell would start selling Quest vWorkspace-based VDI stacks. Not so fast. The vendor partners with Citrix and VMware, so it will continue pointing customers to those companies for VDI software. Still, Dell's recent push in the desktop virtualization market makes it an appealing choice for virtual desktop hardware and software.

Where Eucalyptus stands in the open source cloud battle

For the past two years, experts have had a dim view of Eucalyptus and have thrown support to its open source cloud competitor OpenStack. Recently, however, Eucalyptus Systems has made important strides, positioning itself as an alternative for customers who want to avoid vendor lock-in.

 

With the help of an influx of funding from venture capitalists, the company has progressed at a rapid pace, adding staff and making necessary improvements to Eucalyptus. Among these improvements, the open source cloud vendor dropped its two-license model, simplifying its development and delivery.

 

The release of Eucalyptus 3.1 in June 2012 can be viewed as the completion the platform's restructuring efforts that began in 3.0. The company made substantive changes to improve modularity, allowing for plug-ins and functionality needed beyond its open source code without having to create two development trees. The new architecture makes it possible to develop and deploy plug-ins as separate packages rather than as code that touches multiple packages in a Linux deployment, greatly simplifying development.

 

Eucalyptus 3.1 is almost entirely hosted on Github, simplifying integration with other products and making it easier to build plug-ins and tools. Feature tracking and development activity is publicly available, so any member of the Eucalyptus open source community can request a feature. Bug reports are hosted separately on Jira.

 

New features in Eucalyptus 3.1 include:

• Enterprise platform deployment, which enables enterprise on-premises cloud deployments on the latest version of Red Hat Enterprise Linux (RHEL) with Amazon Elastic Compute Cloud (EC2), Amazon Elastic Block Store (EBS), Amazon Simple Storage Service (S3) and Amazon Identity and Access Management (IAM) using Red Hat Enterprise Virtualization and VMware virtualization platforms.
• FastStart, which provides users with a self-service, automated way to get systems up and running with Eucalyptus 3.1 Infrastructure as a Service (IaaS) in less than 20 minutes.
• Plug-ins and support for paying subscribers. Administrators don't need to modify any code to provide plug-ins, and if you buy the Eucalyptus Enterprise subscription, you get binaries, plug-ins and technical support.
• An automated quality assurance system, which makes it easier to test Eucalyptus 3.1.

 

The Eucalyptus-Amazon relationship

Three months before the release of Eucalyptus 3.1, Eucalyptus Systems formally licensed the Amazon Web Services (AWS) API to build compatible products. This includes APIs for EC2, S3, EBS and IAM. AWS will provide Eucalyptus with API specifications, including related engineering specifications not provided in the public user-level documentation.

 

Eucalyptus has a long-standing relationship with cloud titan AWS, and much of the support for AWS APIs already exists in Eucalyptus -- but this relationship is expected to evolve in Eucalyptus 3.2 and afterward.

 

Many experts view Eucalyptus Systems' relationship with Amazon relationship as a way for the open source cloud provider to take advantage of Amazon's success; however, the partnership has value to AWS, too. The agreement with AWS is about two things: Helping to maintain and enhance its API compatibility in Eucalyptus, and AWS and Eucalyptus Systems going to market together. The agreement can also be viewed as a move to counter the popularity of OpenStack, arguably Eucalyptus' largest competitor.

 

Where's the competition among open source cloud vendors?

There are two major differences between Eucalyptus and OpenStack. First, OpenStack is trying to define its own API, and Eucalyptus is not. Second, OpenStack is a loosely coupled set of components with few use cases in production environments, whereas Eucalyptus is a tested and packaged product. Eucalyptus Systems has three major strengths:

 

• It provides seamless integration with Amazon EC2, the dominant public cloud service, allowing enterprises to move workloads back and forth among Amazon EC2 public clouds and Eucalyptus-based private clouds.
• It supports VMware, allowing customers with VMware virtualized environments to quickly build private clouds.
• It supports all major hypervisors, including VMware ESXi, Xen, Microsoft Hyper-V and KVM.

OpenStack started out with a bang and lots of hype, but it's encountering problems. Issues arise from having too many large vendors involved -- such as Cisco, Dell, HP and Intel -- each with its own agenda. Though some big names, such as HP's Converged Cloud, remain supportive, Citrix, one of the strongest supporters of OpenStack with its Olympus distribution, dropped OpenStack and is focusing on its own open source project, CloudStack.              

 

On the other hand, Eucalyptus is gearing up to become the major open source cloud platform. In April 2012, Eucalyptus Systems was bolstered with another round of funding -- a $30 million investment led by Institutional Venture Partners. The new funding is being used to drive product innovation, sales and customer support. To date, reports state Eucalyptus has raised $55.5 million in capital. And its partner ecosystem has more than doubled to include over 200 of the top cloud and infrastructure automation vendors.

 

Eucalyptus is poised to take advantage of OpenStack's tardiness in moving toward a production-quality cloud technology. However, Rackspace's recent announcement that it was going to begin offering public cloud computing services based on OpenStack could balance the scales. All eyes are on the open source market to see if Eucalyptus 3.1 can propel the vendor to the front of the pack, or if enterprises are still too wary to adopt the technology no matter who the leader is.

 

All good things come to those who wait ... in cloud

The rise of the non-Amazon IaaS cloud provider

Amazon EC2 and Elastic Block Store (EBS) are nearly synonymous with public cloud computing. Since the cloud provider market has matured, enterprises looking for an alternative to the Amazon empire have some other IaaS options.

 

The evolution of the Infrastructure as a Service (IaaS) market is following a predictable pattern. Large-scale vendors that have offered traditional IT services -- IBM, CSC, and Savvis -- also offer general-purpose cloud infrastructure services comparable to Amazon’s. Niche providers also are emerging to offer industry- or regulation-specific services. FireHost, for example, offers PCI-compliant cloud services while Carpathia Hosting targets government services as well as products for the private sector.

 

There also has been innovation in the way in which clouds are managed. Some vendors have dedicated development resources to improve management consoles, other vendors use virtualization platforms that support federation across multiple data centers.

 

The IaaS vendor environment likely will continue to develop; you may want to keep your eye on some of these Amazon alternatives.

 

• BlueLock, a VMware-based virtualization platform, supports hybrid cloud computing architectures. BlueLock is a VMware-certified vCloud Data Center provider.
• CSC, another VMware-certified vCloud Data Center provider, is a well-established IT service provider. The company integrates the Cloud Trust Protocol (CTP) into its offerings. The CTP, which was developed by the Cloud Security Alliance (CSA), enables consumers to query cloud services for information about security configurations and operational characteristics.
• Dell, long known for its servers, desktops and laptops, runs a VMware-based cloud service.
• Even though GoGrid doesn’t have the name recognition of some other IaaS providers, it offers competitive pricing and attractive service-level agreements (SLAs). Developers can use GoGrid’s REST-like application programming interface (API) to maintain control over their virtual devices.
• IBM’s SmartCloud Enterprise product is well suited for test-and-development environments as well as batch jobs. SmartCloud Enterprise is an obvious contender for businesses with a substantial investment in IBM products prior to a cloud initiative.
• Joyent distinguishes itself by offering a tightly integrated software stack in its SmartOS, an operating system designed to support virtualization, storage and performance analytics. SmartOS uses the ZFS storage system DTrace for performance monitoring and a Kernel-based Virtual Machine (KVM) for virtualization.
• Boasting nearly half of Fortune 100 companies as its current customers, Rackspace is major backer of the OpenStack open source platform for cloud computing. Rackspace was founded in 1998 with a focus on hosting services. It has since grown to include nine data centers throughout the U.S., the U.K. and Hong Kong. Rackspace is a major provider in the managed email hosting market.
• Savvis moved into the IaaS market in 2009 with its Savis Cloud Compute service. Its offerings are strong from a security standpoint and support relational databases. The company has 31 data centers worldwide and is based on VMware, making it an attractive option to the financial services industry, businesses requiring PCI compliance as well as other organizations where compliance is an issue.
• Hosting provider Terremark was acquired by Verizon in 2011. The company has data centers in the U. S., Europe and Asia. Cloud consumers can purchase on-demand and dedicated servers from Terremark. The vendor is also a VMware-certified vCloud Data Center provider.
• Windows Azure straddles the line between IaaS and PaaS. Currently, Windows Azure offers VMs as a Web role for IIS-based front-end applications or Work roles that are designed for backend operations. A generic VM role that allows deployment of custom Windows Server 2008 R2 images is in beta.

 

Hedging your bets on a winning IaaS vendor

With such well-known names making their mark on the IaaS market, it’s clear that cloud computing is not a commodity. IaaS providers distinguish themselves with add-on services, 

 

With such a diverse range of products, it’s important to consider which provider best meets your requirements, especially with regard to performance, security and supporting services.

additional support for compliance and integration with existing on-premises infrastructures and application stacks. With such a diverse range of products, it’s important to consider which provider best meets your requirements, especially with regard to performance, security and supporting services.

 

Performance. To ensure high performance in your private cloud, pay particular attention to storage and network services. This is especially important if you’re moving an existing application with a heavy I/O load. Because you won’t have low-level access to the storage systems, some techniques your database administrator and storage admin have used in the past might not be available. Test data transfers between on-premises systems and cloud-based applications with realistic loads to understand how they will function in production.

 

Security. Evaluate the security features a cloud provider offers, including firewall services that may be included at no charge. Other vendors may offer these features at an additional cost. Investigate if intrusion prevention, identity management and other security services are offered and, if so, how they are billed. Some IaaS providers have limited role-based access controls and may not offer the level of granularity you need for access control.

 

Support. In broad terms, support services include automatic backups to customer support. But other management and maintenance terms may be included. Assess what is offered and what those services will cost.

How to build a scalable IaaS cloud network infrastructure

Building an Infrastructure as a Service (IaaS) proof-of-concept is a no-brainer these days—take a few high-end servers, cram as much memory into them as you can, deploy the virtualization platform of your choice (VMware or Xen seem to be the most popular choices), slap some management software in front of the infrastructure—be it a point solution like vCloud Director or a services-oriented one like those offered by Joyent—and finally add a bit of networking and storage to the mix.

 

But scaling the architecture to accommodate an Amazon-sized cloud is a totally different undertaking.

 

Most IaaS architects design their cloud network infrastructure based on two simple—and wrong-headed—assumptions:

• Customers want Layer 2 broadcast domains.
• If you want to achieve high utilization, then having unrestricted virtual machine mobility across the whole infrastructure is a must.

These assumptions unnecessarily limit your design options to stretched VLANs spanning a whole data center. When implemented with traditional data center switches (for example, Cisco Nexus 5000 or equivalent switches from other vendors), the number of VLANs supported by the switches—1,000 to 4,000, depending on the model—directly dictates the number of customers you can support.

 

Amazon Web Services (AWS) has already proven that the customers do not necessarily demand Layer 2 domains in cloud services. AWS doesn’t support broadcasts or IP multicasts, even with its latest set of networking features, but it still boasts tens of thousands of customers. Likewise, unrestricted mobility isn’t a must; you just have to have a large enough number of virtual machines and physical hosts in a cluster to achieve good load distribution. VMware can’t have more than 32 hosts and 3,000 virtual machines in a cluster anyway.

 

Even if your cloud designers insist on having end-to-end VLANs across your data center, all is not lost. After all, service providers have been building large Carrier Ethernet networks for years, and what tenants in your IaaS cloud data center need is no different from what Carrier Ethernet (or Metro LAN if you prefer a better-sounding name) customers need—a VLAN or a set of VLANs between a number of endpoints.

 

Once you get past the notion that you have to buy data center switches, there’s a variety of field-proven choices that you probably already use in your service provider network. You just have to check whether VLAN provisioning on the switches interoperates with your chosen virtualization platform—ideally, the switches would auto-provision VLANs on access trunks based on VMware VirtualCenter information.

 

You can also scale your IaaS cloud infrastructure with judicious use of Layer 3 MPLS VPN services that use VLANs in smaller clusters, or you can use an MPLS/VPN with Virtual Routing and Forwarding (VRF) and MPLS transport in the network core. After all, MPLS/VPN networks have been shown to scale to tens of thousands of customers and hundreds of thousands of routes. There’s no good reason you can’t use the same technology in your data center that you already use in your WAN.

 

Scaling IaaS cloud network infrastructure without workarounds

All of the solutions I’ve mentioned so far, however, are kludges and workarounds caused by the relative lack of networking focus among virtualization vendors. After all, we all know bridging does not scale. And to scale the IaaS cloud, you have to get rid of bridging. Hint: why do you think Amazon can’t offer broadcast or multicast support?

 

Implementing an MPLS/VPN Provider Edge (PE) router in the hypervisor would be one option, but even that doesn’t scale as far as we’d like to go. Every PE router needs a label switched path to every other PE router, which is not a big deal in service provider networks that usually have a few hundred PE routers. Large IaaS data centers could have many more servers—as much as two orders of magnitude more.

 

The only truly scalable solution is IP-over-IP transport since the network core provides scalable IP transport. We know how to design that. After all, we learned a bit while building the Internet, and the hypervisor hosts use the IP core to transport virtual machine data encapsulated in IP datagrams.

 

Sadly, vendors’ hypervisor switches are the weak link

Unfortunately, you can’t buy a hypervisor switch with these features from any virtualization or networking vendor. If you want to be in the same league as Amazon, you have to do what NASA did after the Sputnik launch: figure out what the other side did (after all, the principles are there for all to see), invest a lot in development and build your own system from scratch. Or you could decide that just good enough works for you and build your cloud network infrastructure using the same building blocks as anyone else, but then you have to ask yourself what your competitive advantage will be.

Where Eucalyptus stands in the open source cloud battle

For the past two years, experts have had a dim view of Eucalyptus and have thrown support to its open source cloud competitor OpenStack. Recently, however, Eucalyptus Systems has made important strides, positioning itself as an alternative for customers who want to avoid vendor lock-in.

 

With the help of an influx of funding from venture capitalists, the company has progressed at a rapid pace, adding staff and making necessary improvements to Eucalyptus. Among these improvements, the open source cloud vendor dropped its two-license model, simplifying its development and delivery.

 

The release of Eucalyptus 3.1 in June 2012 can be viewed as the completion the platform's restructuring efforts that began in 3.0. The company made substantive changes to improve modularity, allowing for plug-ins and functionality needed beyond its open source code without having to create two development trees. The new architecture makes it possible to develop and deploy plug-ins as separate packages rather than as code that touches multiple packages in a Linux deployment, greatly simplifying development.

 

Eucalyptus 3.1 is almost entirely hosted on Github, simplifying integration with other products and making it easier to build plug-ins and tools. Feature tracking and development activity is publicly available, so any member of the Eucalyptus open source community can request a feature. Bug reports are hosted separately on Jira.

 

New features in Eucalyptus 3.1 include:

• Enterprise platform deployment, which enables enterprise on-premises cloud deployments on the latest version of Red Hat Enterprise Linux (RHEL) with Amazon Elastic Compute Cloud (EC2), Amazon Elastic Block Store (EBS), Amazon Simple Storage Service (S3) and Amazon Identity and Access Management (IAM) using Red Hat Enterprise Virtualization and VMware virtualization platforms.
• FastStart, which provides users with a self-service, automated way to get systems up and running with Eucalyptus 3.1 Infrastructure as a Service (IaaS) in less than 20 minutes.
• Plug-ins and support for paying subscribers. Administrators don't need to modify any code to provide plug-ins, and if you buy the Eucalyptus Enterprise subscription, you get binaries, plug-ins and technical support.
• An automated quality assurance system, which makes it easier to test Eucalyptus 3.1.
• The Eucalyptus-Amazon relationship
• Three months before the release of Eucalyptus 3.1, Eucalyptus Systems formally licensed the Amazon Web Services (AWS) API to build compatible products. This includes APIs for EC2, S3, EBS and IAM. AWS will provide Eucalyptus with API specifications, including related engineering specifications not provided in the public user-level documentation.

 

Eucalyptus has a long-standing relationship with cloud titan AWS, and much of the support for AWS APIs already exists in Eucalyptus -- but this relationship is expected to evolve in Eucalyptus 3.2 and afterward.

 

Many experts view Eucalyptus Systems' relationship with Amazon relationship as a way for the open source cloud provider to take advantage of Amazon's success; however, the partnership has value to AWS, too. The agreement with AWS is about two things: Helping to maintain and enhance its API compatibility in Eucalyptus, and AWS and Eucalyptus Systems going to market together. The agreement can also be viewed as a move to counter the popularity of OpenStack, arguably Eucalyptus' largest competitor.

 

Where's the competition among open source cloud vendors?

There are two major differences between Eucalyptus and OpenStack. First, OpenStack is trying to define its own API, and Eucalyptus is not. Second, OpenStack is a loosely coupled set of components with few use cases in production environments, whereas Eucalyptus is a tested and packaged product. Eucalyptus Systems has three major strengths:

 

It supports all major hypervisors, including VMware ESXi, Xen, Microsoft Hyper-V and KVM.

OpenStack started out with a bang and lots of hype, but it's encountering problems. Issues arise from having too many large vendors involved -- such as Cisco, Dell, HP and Intel -- each with its own agenda. Though some big names, such as HP's Converged Cloud, remain supportive, Citrix, one of the strongest supporters of OpenStack with its Olympus distribution, dropped OpenStack and is focusing on its own open source project, CloudStack.              

 

On the other hand, Eucalyptus is gearing up to become the major open source cloud platform. In April 2012, Eucalyptus Systems was bolstered with another round of funding -- a $30 million investment led by Institutional Venture Partners. The new funding is being used to drive product innovation, sales and customer support. To date, reports state Eucalyptus has raised $55.5 million in capital. And its partner ecosystem has more than doubled to include over 200 of the top cloud and infrastructure automation vendors.

 

Eucalyptus is poised to take advantage of OpenStack's tardiness in moving toward a production-quality cloud technology. However, Rackspace's recent announcement that it was going to begin offering public cloud computing services based on OpenStack could balance the scales. All eyes are on the open source market to see if Eucalyptus 3.1 can propel the vendor to the front of the pack, or if enterprises are still too wary to adopt the technology no matter who the leader is.

 

Cloud development tools to keep on your radar

Most public cloud announcements focus on reduced rates for machine instances, new persistent storage options or the introduction of new services, while news about cloud development tools seems to get the short shrift. Still, Amazon and Microsoft offer many new tools that make it easier for developers to build, test and deploy instances in the cloud.

Developers looking for language-specific cloud tools to help streamline some of the more tedious cloud-instance and storage tasks have several options. Let's take a look at what cloud development tools Microsoft and Amazon have to offer.

Amazon Toolkit for Eclipse supports cloud development
Amazon is focusing much of its developer support efforts on Eclipse, the widely used developer framework. Amazon has created a toolkit that supports Java developers as well as any who work with its cloud infrastructures, which include Elastic Compute Cloud (EC2), Amazon Simple Storage Service (S3), Amazon SimpleDB and others. Developers can download the AWS Toolkit for Eclipse from the AWS website; Eclipse is available from the Eclipse project website.

After installing the AWS Toolkit for Eclipse and configuring it with your AWS access key and secret key, you can work with the preinstalled supported services. The SDK includes example applications for working with S3, SimpleDB, Simple Queue Service (SQS) and the AWS console. Instructions how to get started with AWS SDK for Java are available online.

In addition to writing code, developers need to be able to manage their resources. Amazon Toolkit for Eclipse allows developers to perform maintenance tasks in the same development environment they're coding in.

The AWS SDK and AWS Explorer let developers manage S3, SQS, relational databases (using the Relational Data Services), EC2 instances and the AWS Elastic Beanstalk. The toolkit supports multiple Amazon accounts, so you can configure which account to use by default.

Explorer implements functions specific to each supported Amazon web service. For example, in the S3 section of AWS Explorer, you can list buckets, delete buckets and change permissions, among other things. In its Simple Queue Service editor, developers can list, delete and send messages to queues as well as delete messages in your queues.

The query editor in the SimpleDB feature allows cloud developers to find data that meets specific criteria. The Amazon EC2 tool lets developers perform basic tasks on machine images, instances, elastic block storage and security groups. When it's time to deploy a cloud application to a server, the AWS Elastic Beanstalk tool in Explorer can streamline the task.

Microsoft's cloud development tools
Microsoft is doing its part to support developers and can work with numerous programming languages, including .NET, node.js, Java, PHP, Python and others. Microsoft's embrace of multiple languages extends to non-Microsoft platforms as well. For example, the company includes a tutorial on using Python, MySQL and Django, a Python framework for building websites. Additionally, it's not surprising that there is close integration between Microsoft Windows Azure and Visual Studio 2012.

Microsoft is clearly working to support developers with a number of common languages. Visual Studio supports a server explorer that allows you to view information about server configurations while developing. For example, you can check on the status of your virtual machines (VMs) and connect to them by using Remote Desktop within the development environment. You can also manipulate additional services within Windows Azure, including establishing a caching service for your application within Visual Studio. A recent release of Visual Studio allows developers to deploy simultaneous updates to multiple instances -- a potential timesaver for distributed applications such as cloud apps.

Visual Studio includes features that help with common development tasks as well as more targeted requirements. Microsoft has added additional support via NuGet, a tool for adding third-party libraries to a Visual Studio environment with a focus on runtime and diagnostic and configuration packages. Visual Studio includes templates for common functional roles, such as Worker Role with Service Bus Queue and Cache Worker role.

Whether you prefer either Amazon's or Microsoft's cloud for cloud development work, both have advantages. Amazon's emphasis is on Java developers, but it does offer support for a range of development tasks -- from managing VM instances to querying data.

Microsoft is clearly working to support developers with a number of common languages. Its flagship development platform, Visual Studio, is keeping pace with changing needs as developers move to the cloud.

Thought VMware vRAM licensing was dead? Think again...

VMware isn't completely "striking the word vRAM from the vocabulary of the vDictionary," as CEO Pat Gelsinger said at VMworld. The controversial pricing model will still apply to those running vSphere in public clouds. Participants in the VMware Service Provider Program (VSPP) will continue to charge according to the amount of memory allocated per virtual machine (VM).

Some service providers said they have no major problems with vRAM licensing and pricing because it aligns well with the public cloud's pay-per-use business model. "We're fairly happy with where we are right now," said Pat O'Day, chief technology officer of VSPP partner BlueLock LLC. But a West Coast-based VSPP partner said VMware vRAM can put VSPPs at a disadvantage against other service providers, especially when it comes to high-density hosts. "We're seeing a lot of large service providers and industry players moving off of VMware and into open source … and that's putting competitive pressure on us," said the partner, who spoke on the condition of anonymity.

How VMware vRAM works for VSPPs
The VMware vRAM allocation limits for VSPPs are different from those abolished at VMworld, which were for business customers purchasing vSphere. For VSPPs, vRAM is capped at 24 GB of reserved memory per VM, and VSPPs are charged for at least 50% of the memory they reserve for each VM.

Also, unlike with the licenses for private organizations, VSPP vRAM is not associated with a CPU-based entitlement, and it does not require the purchase of additional licenses to accommodate memory pool size limits.

The infrastructure for BlueLock Virtual Datacenters runs on Hewlett-Packard servers, packing in 512 GB of RAM each. VMware licensing is a very small percentage of BlueLock's overall costs, O'Day said. Given the 24 GB per VM cap, the company could probably add even more RAM to each server, O'Day said. "But that's not worth disrupting our business model to do," he added.

VSPPs' VMware vRAM reactions
VMware discontinued the wildly unpopular vRAM licensing and pricing model for business customers after more than a year of discontent. Some VSPPs worried the move away from vRAM on the enterprise side would also affect their businesses, according to an email one VSPP software distributor sent to service provider clients.
"We have received feedback from quite a few partners who have expressed concern to us over the announcement that VMware is moving away from vRAM," the distributor said in the email. "The question to a person seems to be, 'How does this impact me as a VSPP partner whose billing is completely based around vRAM?' The answer is, not at all."

But the West Coast partner, who received that email, said the cost of memory on high-density hosts can eat into VSPPs' margins or force VSPPs to pass the costs on to end-user customers. This partner offers services including hosted virtual desktops that don't run on VMware View, and vRAM can cost as much as $7 per gigabyte under this model.

"If you're trying to provision a desktop with four gigabytes of RAM … the memory alone is $30," the partner said. "You're at a significant disadvantage as the market tries to bring the price down."

Furthermore, the complexity of the licensing model makes it difficult to set the monthly costs of vRAM based on usage, the partner said.

"You basically hand VMware your wallet and let them take whatever they want," the partner added. "… The pricing is so complicated, we have no way to estimate how much we're going to pay VMware at the end of the month."

In a blog post discussing the continuation of vRAM for VSPPs, VMware said the vRAM model allows service providers to sell more computing capability from the same infrastructure, and to control the memory oversubscription and service profit margin through allocated vRAM delivered to customers.