Posts tagged economies of scale
The Many Benefits of Cloud Computing
Jun 10th
Cloud computing brings benefits across three categories: economic, architectural, and strategic. First, it helps reduce IT costs and shift budgets from CAPEX to OPEX. Second, it improves the experience of end-users. Third, it helps companies focus on their core competencies.
Utility Pricing
Cloud Computing is first a foremost defined by its utility-based pricing model. Users of the platform consume computing and storage services on demand and pay for them as they go, using an Operating Expenses (OPEX) budget, instead of paying for infrastructure resources up-front using Capital Expenditures (CAPEX). For example, a Director of Sales can create user accounts for 10 of her sales people on some on-demand CRM application by using her corporate credit card, without having to ask the CFO for a budget, and without having the IT Department initiate a requisition process for a new server.
Elastic Resource Capacity
Cloud Computing differs from more traditional forms of distributed computing in the way it scales computing and storage resources up and down. Instead of tapping from a fixed set of resources, users can add or remove capacity at will, almost instantaneously, and only pay for what they actually use. While utility pricing let users pay as they go, elastic resource capacity let them pay as they grow (or shrink). Following our previous example, the Director of Sales can add 5 more accounts for the sales people that were recently added to her team following the merger with another company, without having to worry about adding new servers or buying more hard drives.
Virtualized Resources
Cloud Computing would not be possible without virtualization, not for arcane technical reasons, but for one obvious business requirement: the need for multi-tenancy. In order to benefit from economies of scale, cloud computing is predicated upon the sharing of a common infrastructure by multiple groups of users, often referred to as tenants. And multi-tenancy can only be achieved through some kind of virtualization, either at the database level (Salesforce.com), application server level (Google AppEngine), kernel level (Red Hat), or CPU level (Amazon EC2). Unlike grid computing, which often pooled and aggregated distributed computing resources for the purpose of handling very large computing jobs that could not fit or would take too long to complete on a single server, Cloud Computing creates virtual slices of resources from clusters of servers and storage devices, perfectly sized to fit the specific needs of multiple users. Such virtual resources can be small or large, and scale elastically as user needs evolve over time. In our previous example, virtualization means that the CRM application used by our sales team is served by an infrastructure also used by many other tenants, all securely isolated from each other.
Management Automation
Cloud Computing platforms differ from traditional corporate data-centers in one major way: standardization. While your typical data-center will usually host every versions of every operating systems and databases known to man, thereby creating massive management overhead, most Cloud Computing platforms usually standardize on a single kind of CPU (x86-based predominantly), a single hypervisor (VMware, Xen, etc.), a single operating system (some Linux distribution usually), and a single database (MySQL is the dominant player there). This standardization has an obvious business benefit: dramatic reduction of operating costs through aggressive management automation. Following our previous example, the sales team’s CRM application is served by a single infrastructure managed by a team that is orders of magnitude smaller than the aggregated team that would be required to manage the systems used by all tenants if every one of them were to use their own infrastructure.
Self-service Provisioning
Cloud Computing and Software as a Service are often compared to the Application Service Provider (ASP) model that became popular for a brief period of time in the late 90′s. One element makes them fundamentally different from each other though: self-service provisioning. With the ASP model, dedicated servers had to be provisioned for every customer, which meant that technical resources had to be involved every time a new customer would be signed. Hefty setup fees would be added to the bill, and the service would become operational within a few days at best. With Cloud Computing, business end-users like our Director of Sales can provision applications and user accounts in a few mouse clicks, and these become available instantly.
Third-party Ownership
Cloud Computing is also a new form of outsourcing. Customers trying to focus the allocation of scarce capital resources to their core businesses soon realize the benefits of moving IT infrastructure off their balance sheet. Furthermore, as technology evolves and leading service providers roll-out ever larger data-centers, the acquisition and operation of state-of-the-art data-center facilities makes less and less sense from an economic standpoint for most organizations. Cloud Computing is all about the transfer of ownership for such resources to a third-party that specializes in their deployment. According to our previous example, the company using the CRM application managed by some service provider does not own any infrastructure beyond a few laptop computers. Everything else, from data-centers to servers and storage systems is owned and operated by the service provider.
Managed Operations
Cloud Computing is finally about allocating human resources to tasks that will directly impact the business, rather than simply managing the infrastructure that supports it. As such, Cloud Computing advocates a model according to which the IT infrastructure is not only owned by a third-party, but managed by the third-party as well. Software upgrades, data backups, and the countless other tasks required to manage mission-critical business applications on a day to day basis become the responsibility of a third-party, according to well-defined Service Level Agreements. Following our example, the Director of Sales discovered this morning the umbrella adorned logo for the Summer 2010 version of her favorite CRM application, without having taken any part in the software upgrade process that happened over the weekend. In the cloud, ignorance is bliss.
Article Contribution to CloudTweaks.com by: Ismael Chang Ghalimi
About Ismael
Ismael Ghalimi is a passionate entrepreneur and fervent industry observer, founder and CEO of Intalio.
More information on Ismael can be found at: Intalio
Related posts
Why Cloud Computing Will Never Be Free
May 27th
The last time the IT industry delivered outsourced shared-resource computing to the enterprise was with timesharing in the 1980s when it evolved to a high art, delivering the reliability, performance, and service the enterprise demanded. Today, cloud computing is poised to address the needs of the same market, based on a revolution of new technologies, significant unused computing capacity in corporate data centers, and the development of a highly capable Internet data communications infrastructure. The economies of scale of delivering computing from a centralized, shared infrastructure have set the expectation among customers that cloud computing costs will be significantly lower than those incurred from providing their own computing. Together with the reduced deployment costs of open source software and the perfect competition characteristics of remote computing, these expectations set the stage for fierce pressure on cloud providers to continuously lower prices.
This pricing pressure results in a commoditization of cloud services that deemphasizes enterprise requirements such as guaranteed levels of performance, uptime, and vendor responsiveness, much as has been the case with the Web hosting industry. Notwithstanding, it is the expectation of enterprise management that operating expenses be reduced through the use of cloud computing to replace new and existing IT infrastructure. This difference between expectation and what the industry can deliver at today’s nearzero price points represents a challenge, both technical and organizational, which will have to be overcome to ensure large-scale adoption of cloud computing by the enterprise.
The Essential Characteristics of Cloud Computing
This is where we come full circle and timesharing is reborn. The same forces are at work that made timesharing a viable option 30 years ago: the high cost of computing (far exceeding the cost of the physical systems), and the highly specialized labor needed to keep it running well. The essential characteristics of cloud computing that address these needs are:4
- On-demand access. Rapid fulfillment of demand for computing and continuing ability to fulfill that demand as required.
- Elasticity. Computing is provided in the amount required and disposed of when no longer needed.
- Pay-per-use. Much like a utility, cloud resource charges are based on the quantity used.
- Connectivity. All of the servers are connected to a high-speed network that allows data to flow to the Internet as well as between computing and storage elements.
- Resource pooling. The cloud provider’s infrastructure is shared across some number of end customers, providing economies of scale at the computing and services layers
- Abstracted infrastructure. The cloud end customer does not know the exact location or the type of computer(s) their applications are running on. Instead, the cloud provider provides performance metrics to guarantee a minimum performance level.
- Little or no commitment. This is an important aspect of today’s cloud computing offerings, but as we will see here, interferes with delivery of the services the enterprise demands.
Continue… Credit to: Communications of the ACM
Related posts
Cloud Computing – Understanding Cloud Datacenter Economies of Scale
May 4th
I was quite impressed by James Hamilton’s recent MIX’10 presentation on economies of scale for large cloud providers. James “gets it” like few others in the industry. If you haven’t watched his hour-long presentation, I suggest you do. I also recommend this excellent response from James Urquhart. My goal in this posting is to highlight, clarify and expand on a few of James Hamilton’s points. I will focus on Infrastructure-as-a-Service (IaaS) clouds, but the concepts are relevant for other kinds of cloud services.
In his presentation, James focuses on power: utilization, distribution, etc., and while an important element, like him, I don’t think it’s the most important factor.
I also want to dispel the myth that only the largest companies can achieve these economies of scale. Don’t get me wrong; providing a cloud service is a scale game. It requires a certain amount of buying power to compete. However,you don’t need to be MSFT, YHOO, AMZN, or GOOG to compete effectively. Buying power can be had at levels much lower than you might think.
In this article, I refer regularly to Jame’s comments in his presentation, so I suggest you watch his video first. In order to minimize confusion, I’ve borrowed some pictures from his slides and inserted them here for your reference. This is a long entry, but it will be worth the read as I’ve got numbers for you which I hope you will find interesting.
Background
Like James, the Cloudscaling team has a history of building large scale services. I’ve worked in this area for 16+ years as has our COO, Adam Waters, and several of our team members. Understanding of the economies of scale, especially for service providers, cloud or otherwise is fundamental to our DNA. For example, see my previous piece describing how oversubscription works.
Enough of that! Let’s dig in and look at where you can achieve economies of scale, identifying areas James Hamilton may have neglected, and clarifying areas where I think there is still confusion.
Economies of Scale
There are a number of areas where you can achieve economies. James touched on a few in his talk. While this is not a 100% complete list, here are key areas of opportunity that I see:
- Datacenter and facilities architecture (power & cooling)
- Buying power (COGS) for Networking, Compute (Servers), and Storage
- Development & Labor Costs
- Standardization & Homogenization
- Cash Flow
In James Hamilton’s model (see pie chart below) server costs are the dominant cost, but he critically left out development & labor costs. This can be as much as 10% for a cloud and while it’s possible for large clouds to drive this down to a marginal cost, in practice there are no Infrastructure-as-a-Service clouds of sufficient size to achieve this yet. While James focuses primarily on power & cooling in his presentation, let’s take a closer look at some other areas.
Jame’s Hamilton’s Distribution of Cloud Datacenter Costs
Networking
There are two key areas of networking where you can achieve economies of scale:
- Buying IP (network) transit (OpEx)
- Capital expense (CapEx)
Related posts
Cloud Computing: It’s the Economics Stupid
Apr 23rd
Cloud’s ability to make IT more agile is undeniable, says CIO.com’s Bernard Golden. But once that benchmark is met, the cost effectiveness of the internal cloud versus public providers will be a key measure of success.
CIO — The question of costs associated with cloud computing continue to be controversial. You may recognize in this blog’s title, an homage to the motto of Bill Clinton’s 1992 Presidential campaign: “It’s the Economy, Stupid.” The motto referred to the decision by the Clinton campaign to focus relentlessly on how the U.S. economy was doing in 1992, sidestepping other issues and always, always circling back to the economic outlook for the US. I was reminded of this by some recent discussions on Twitter that discussed the importance of economics in terms of cloud adoption.
This question of cloud economics arises especially in the context of the endless discussions about private vs. public clouds (private usually being thought of as referring to a cloud environment inside a company’s own data center). Some people assert that private clouds obviously must be less expensive, because one owns the equipment and is not paying what is, in effect, a rental fee. The obvious analogy is buying a car vs. renting a car. If one uses a car every day, it’s clearly less expensive to own rather than pay a daily rental fee to, say, Hertz. Sometimes this argument is made stronger by noting that public cloud providers also are profit-seeking enterprises, so an extra tranche of end-user cost is present, representing the profit margin of the public offering.
The proponents of public cloud computing cost advantages point to the economies of scale large providers realize. At a recent “AWS in the Enterprise” event, Werner Vogels, CTO of Amazon, noted that Amazon buys “10s of racks of servers at a time” and gets big discounts because of this. Also, AWS buys custom-designed equipment that leave out unneeded, power-using features like USB ports. Moreover, the public cloud providers implement operations automation to an extreme degree and thereby drop the labor cost factor in their clouds.
Related posts
Saving Money Through Cloud Computing – Cloud Information
Apr 7th
The U.S. federal government spends nearly $76 billion each year on information technology, and $20 billion of that is devoted to hardware, software, and file servers (Alford and Morton, 2009). Traditionally, computing services have been delivered through desktops or laptops operated by proprietary software. But new advances in cloud computing have made it possible for public and private sector agencies alike to access software, services, and data storage through remote file servers. With the number of federal data centers having skyrocketed from 493 to 1,200 over the past decade (Federal Communications Commission, 2010), it is time to more seriously consider whether money can be saved through greater reliance on cloud computing.
Cloud computing refers to services, applications, and data storage delivered online through powerful file servers. As pointed out by Jeffrey Rayport and Andrew Heyward (2009), cloud computing has the potential to produce “an explosion in creativity, diversity, and democratization predicated on creating ubiquitous access to high-powered computing resources.” By freeing users from being tied to desktop computers and specific geographic locations, clouds revolutionize the manner in which people, businesses, and governments may undertake basic computational and communication tasks (Benioff, 2009). In addition, clouds enable organizations to scale up or down to the level of needed service so that people can optimize their needed capacity. Fifty-eight percent of private sector information technology executives anticipate that “cloud computing will cause a radical shift in IT and 47 percent say they’re already using it or actively researching it” (Forrest, 2009, p. 5).
To evaluate the possible cost savings a federal agency might expect from migrating to the cloud, in this study I review past studies, undertake case studies of government agencies that have made the move, and discuss the future of cloud computing. I found that the agencies generally saw between 25 and 50 percent savings in moving to the cloud. For the federal government as a whole, this translates into billions in cost savings, depending on the scope of the transition. Many factors go into such assessments, such as the nature of the migration, a reliance on public versus private clouds, the need for privacy and security, the number of file servers before and after migration, the extent of labor savings, and file server storage utilization rates. Based on this analysis, I recommend five steps be undertaken in order to improve efficiency and operations in the public sector:
- the government needs to redirect greater resources to cloud computing in order to reap efficiencies represented by that approach,
- the General Services Administration should compile data on cloud computing applications, information storage, and cost savings in order to determine possible economies of scale generated by cloud computing,
- officials should clarify procurement rules to facilitate purchasing through measured or subscription cloud services and cloud solutions appropriate for low, medium, and high-risk applications,
- countries need to harmonize their laws on cloud computing to avoid a “Tower of Babel” and reduce current inconsistencies in regard to privacy, data storage, security processes, and personnel training, and
- lawmakers need to examine rules relating to privacy and security to make sure agencies have safeguards appropriate to their mission.
