25-03-2014, 19:39 #1
[EN] Google Slashes Cloud PricingThe 800 pound Gorilla of cloud computing has awoken. Google today demonstrated some cool new developer-friendly features and announced major price drops for its Google Cloud Platform services.
At the Google Cloud Platform live event today, Google senior vice president Urs Hölzle demonstrated Live Migration, a feature which allows customers to seamlessly move virtual machines between data centers without interrupting service, and could be enormously useful in allowing users to route around data center outages.
The company also introduced CloudDNS, as well as added a command line tool and Managed VMs, which blur the line between Platform as a Service and Infrastructure as a service, allowing developers to leverage the best of both worlds.
Google also added support for VMs using the Windows operating system, something that has long been in demand by customers. Preview support of Windows Server 2008 R2 is available starting today. The company also announced general availability of Linux operating systems from SUSE and RedHat Linux on their cloud.
Cloud Pricing Meets Moore’s Law
“Pricing hasn’t followed Moore’s Law: over the past five years, hardware costs improved by 20-30 percent annually but public cloud prices fell at just 8 percent per year,” Hölzle wrote in a blog post summarizing the announcements. “We think cloud pricing should track Moore’s Law, so we’re simplifying and reducing prices for our various on-demand, pay-as-you-go services by 30-85%:
Google will have on-demand price reductions as well as sustained-use discounts, starting with a 32 percent price drop today. Storage dropped a staggering 68 percent, at .026 per GB, or .02 per GB DRA. BigQuery prices dropped 85 percent.
“The Google cloud platform is a central part of our infrastructure development, and we’re investing heavily in it to make it as great a platform to external users as it has been to internal users at Google,” said Hölzle, who said Google has laid the groundwork for years and years of future improvements. The aim is to make developers more productive.
New Pricing Takes Guesswork Out
With Google’s new pricing, the on-demand price for VMs is now lower than the three-year reserve for most providers. Hölzle said there is too much complexity optimizing cost and performance. Hölzle says that current clouds force too many trade offs.
“Pricing is still way too complex,” said Hölzle. “It seems like you need a PhD to figure out the best option.”
That was reinforced by an analysis from cloud integrator RightScale that compared Google’s new pricing to cloud market leader Amazon Web Services, which allows users to manage future capacity through the ourchase of reserved instances.
“The new Google sustained-use pricing avoids the complexity, lock-in, and upfront costs of AWS reserved instance purchases,” RightScale notes. “Google users will automatically receive the best price for their level of usage, with no planning required on their part. However, tying the sustained-use discounts to Google’s monthly billing cycle could create an incentive to make decisions such as switching instance sizes on the monthly billing boundaries.”
This kind of pricing is tricky for customers. Reserved instances have a purpose: they try to reward you for predictable sustained workload. With Google’s approach to pricing, a customer’s price begins to drop once they use a VM for more than 25 percent of the month, equating to a 53 percent discount over today’s prices. This applies to all instances of a certain type in a region. As long as the VM in aggregate is used more than 25 percent, customers get the discount.
Live Migration Is Seamless
One big feature is Live Migration in Google Compute Engine. “As a cloud provider, you’re faced with a dilemma,” said Hölzle. “Because you can either have an up to date infrastructure – patch operating systems, upgrade, introduce new features and so on at a rapid pace – or you can provide long-running VMs to users. But you can’t really do both. Live Migration solves this dilemma and lets us provide both.”
Google’s Chris Elliot gave a demo of Live Migration to show how seamless it is for customers. Elliot moved an application to a new physical machine live. The instance flashed yellow, signaling its move to a new physical host. An engineer in Seattle simulated a physical hardware failure, which triggered the live migration. 1080p video streamed while the instance was undergoing live migration. The migration occurred with no effect to the video.
The company can live migrate customers around hardware issues, giving customers the safest environment to work on. It has added intelligence to that layer It is completely transparent; customers can’t tell when live migration occurs and work normally. The only way they know is a little note in the operations log telling them of the migration. It also allows
Google to upgrade infrastructure in the background with no negative effects to those hosted on its cloud. The instance retains the same IP and same configuration, only the physical machine has changed.
“There is a lot of innovation under the covers,” said Hölzle.
Improvements to BigQuery
On the Big Data front, Google added predictable throughput for BigQuery, its tool for running SQL-like analytic queries against multi terabyte database. You can analyze any amount of data at 5 gigabyte increments with a chosen level of throughput at a fixed price. Compared to current industry pricing, it’s 75 percent less expensive.
The company also introduced DNS as a service, built and operated by Google, with a simple API and integrated into the Google console. You can now manage your network configurations and DNS in the same place.
The company has also blended PaaS and IaaS. Managed Virtual Machines Combines the freedom of the IaaS Compute engine, with the capabilities of the PaaS AppEngine. With just a few configuration line changes, you can take an AppEngine machine out and use it in Compute Engine.
A Look Back
A year ago in May Google Compute Engine went into preview, as well as PHP support for App Engine. The company added Layer 3 Load balancing in August. Demoed in November, it can handle a million requests per second from a cold start. It’s based on the same codebase that powers load balancing for google.com
In December, Compute Engine went into General Availability. The company then introduced cloud datastore, based on BigTable, the system that start noSQL revolution. In August, the company added encryption at rest for Google Cloud storage. Cloud SQL went into General availability last month.Offline disk import was added in June 2013.
The company has made improvements to BigQuery, added Dedicated Memecache, and added a mobile backend starter for iOS. This makes it easier to build applications for Apple phones in addition to Android. In December, it added Persistent Disks, and last month it released Cloud SQL into general availability, with a 99.95 percent Service Level Agreement. HIPAA support was added in February of this year as well.
25-03-2014, 19:48 #2
Google’s Bigger, Cheaper CloudGoogle’s cloud computing business has figured out how it’s going to come at Amazon Web Services: lower and simpler prices, predictable services and software innovation. Some of the price cuts are as much as 85 percent. Amazon is expected to soon respond with moves of its own.
Over the long term, according to the executive in charge of Google’s public cloud, Google wants software developers, both independent and within companies, to count on prices steadily dropping at a rate that roughly tracks Moore’s Law. That means a doubling of power per dollar every 18 to 24 months. Google also wants its cloud business to move from a series of different services to a single, comprehensive product that handles familiar business applications, new apps, and big back-end data analysis and storage jobs.
“We think if we win the hearts and minds of developers, good things happen,” said Greg DeMichillie, director of management for Google’s cloud business. “Simplicity will be a big deal – what developers like to do is to write code, not managing infrastructure or worrying about resource planning.”
The lower pricing and the predictability are smart moves, strategically. While Amazon Web Services has marquee names like Netflix and Shell renting its sophisticated computers over the Internet to run businesses, there is a sense among many smaller companies that the Amazon service is a good place to get started, but it becomes expensive as a company gets bigger. For example, the business intelligence company Good Data was once an advocate of Amazon Web Services, but has moved most operations into its own cloud system.
“No one should ever say it’s cheaper to run this themselves,” Mr. DeMichillie said. A reliable drop in prices could also make it easier for companies to plan the creation of new products and services. Amazon Web Services has often cut prices aggressively, but in something of a scattered way.
At a Google event on Tuesday, Urs Hölzle, a Google senior vice president, said prices for Google Compute Engine, its cloud service for big workloads, would be cut 32 percent across the board. Prices for App Engine, its software application service, were simplified and fell about 30 percent. Data storage prices were cut 68 percent, in general, to 2.6 cents per gigabyte over various storage systems. BigQuery, a data analysis product, saw prices cut 85 percent.
Prices would continue to fall roughly in line with Moore’s Law, Mr. Hölzle said. That is actually an impressive vow; Moore’s Law is an observation about semiconductor power, but Google, Amazon Web Services and others offer increasingly sophisticated software, as well as hardware. Google has internal methods of creating and deploying systems, Mr. DeMichillie said, that make it confident it can make good on the pricing vow.
“Our velocity can be made faster than the industry,” he said. As to whether anyone in Google had qualms about exposing its crown jewels, he said, “there is a recognition throughout Google that cloud is the next big thing. There is no sense of ‘No, not that.’”
The announcement was made at a Google cloud event in San Francisco. On Wednesday, Amazon Web Services will have its own event in San Francisco and may respond to Google’s moves with announcements of its own.
Sustained users of Google’s cloud, as opposed to intermittent purchasers, would receive discounts of up to 30 percent.
In addition to the lower prices, Google added a couple of features, such as the ability for a software developer to rapidly look as source code, automatic updating of software that is in the process of completion, and minor code changes while viewing it with a browser, that can speed production of new software.
Future product announcements this year, Mr. DeMichillie said, will involve “batch computing, opportunistic computing – where you get 100,000 computing cores for a brief period.” IBM and Amazon Web Services have been involved in similar initiatives.
Google is also allowing operating systems such as Suse, Red Hat Linux and Microsoft Windows Server to be used within its cloud. This is considered a minimum necessity for many established businesses. BigQuery is being modified to take in 100,000 data records per second, allowing for the fast analysis of large streams of data.
The many developer-friendly improvements in Google’s Cloud make somewhat conspicuous through absence what has long been a weak spot where Google’s commercial side is concerned: While its geek-heavy culture speaks to developers, the company has had less success understanding and speaking to more conventional businesspeople. While now a $1 billion a year business, Google Apps for Business struggled for many years to make significant progress against Microsoft Office, despite being significantly cheaper.
Mr. DeMichillie said that is changing too. “We are good and getting better,” he said. “Of course we talk to executives. It’s not like the start-ups were beating down our door to have Windows Server.”
25-03-2014, 19:51 #3
Google Cloud Platform Live - Blending IaaS and PaaS, Moore’s Law for the cloudThe original promise of cloud computing was simple: virtualize hardware, pay only for what you use, with no upfront capital expenditures and lower prices than on-premise solutions. But pricing hasn’t followed Moore's Law: over the past five years, hardware costs improved by 20-30% annually but public cloud prices fell at just 8% per year.
We think cloud pricing should track Moore’s Law, so we’re simplifying and reducing prices for our various on-demand, pay-as-you-go services by 30-85%:
- Compute Engine reduced by 32% across all sizes, regions, and classes.
- App Engine pricing is drastically simplified. We've lowered pricing for instance-hours by 37.5%, dedicated memcache by 50% and Datastore writes by 33%. In addition, many services, including SNI SSL and PageSpeed are now offered to all applications at no extra cost.
- Cloud Storage is now priced at a consistent 2.6 cents per GB. That’s roughly 68% less for most customers.
- Google BigQuery on-demand prices reduced by 85%.
In addition to lower on-demand prices, you’ll save even more money with Sustained-Use Discounts for steady-state workloads. Discounts start automatically when you use a VM for over 25% of the month. When you use a VM for an entire month, you save an additional 30% over the new on-demand prices, for a total reduction of 53% over our original prices.
With our new pricing and sustained use discounts, you get the best performance at the lowest price in the industry. No upfront payments, no lock-in, and no need to predict future use.
Making developers more productive in the cloud
We’re also introducing features that make development more productive:
- Build, test, and release in the cloud, with minimal setup or changes to your workflow. Simply commit a change with git and we’ll run a clean build and all unit tests.
- Aggregated logs across all your instances, with filtering and search tools.
- Detailed stack traces for bugs, with one-click access to the exact version of the code that caused the issue. You can even make small code changes right in the browser.
We’re working on even more features to ensure that our platform is the most productive place for developers. Stay tuned.
Introducing Managed Virtual Machines
You shouldn't have to choose between the flexibility of VMs and the auto-management and scaling provided by App Engine. Managed VMs let you run any binary inside a VM and turn it into a part of your App Engine app with just a few lines of code. App Engine will automatically manage these VMs for you.
Expanded Compute Engine operating system support
We now support Windows Server 2008 R2 on Compute Engine in limited preview and Red Hat Enterprise Linux and SUSE Linux Enterprise Server are now available to everyone.
Real-Time Big Data
BigQuery lets you run interactive SQL queries against datasets of any size in seconds using a fully managed service, with no setup and no configuration. Starting today, with BigQuery Streaming, you can ingest 100,000 records per second per table with near-instant updates, so you can analyze massive data streams in real time. Yet, BigQuery is very affordable: on-demand queries now only cost $5 per TB and 5 GB/sec reserved query capacity starts at $20,000/month, 75% lower than other providers.
This is an exciting time to be a developer and build apps for a global audience. Today we’ve focused a lot on productivity, making it easier to build and test in the cloud, using the tools you’re already familiar with. Managed VMs give you the freedom to combine flexible VMs with the auto-management of App Engine. BigQuery allows big data analysis to just work, at any scale.
And on top of all of that, we’re making it more affordable than it’s ever been before, reintroducing Moore’s Law to the cloud: the cost of virtualized hardware should fall in line with the cost of the underlying real hardware. And you automatically get discounts for sustained use with no long-term contracts, no lock-in, and no upfront costs, so you get the best price and the best performance without needing a PhD in Finance.
We’ve made a lot of progress this first quarter and you’ll hear even more at Google I/O in June.
-Posted by Urs Hölzle, Senior Vice President
25-03-2014, 19:57 #4
Google Announces Massive Price Drops For Its Cloud Computing Services And Storage
At its Cloud Platform event in San Francisco today, Google announced a number of massive price drops for virtually all of its cloud-computing services. The company has also decided to reduce the complexity of its pricing charts by eliminating some charges and consolidating others.
Google Compute Engine is seeing a 32 percent reduction in prices across all regions, sizes and classes. App Engine prices are down 30 percent, and the company is also simplifying its price structure.
The price of cloud storage is dropping a whopping 68 percent to just $0.026/month per gigabyte and $0.2/month per gigabyte/DRA. At that price, the new pricing is still lower than the original discount available for those who stored more than 4,500TB of data in Google’s cloud.
The prices for other cloud storage options, including Cloud SQL and the Cloud Datastore, aren’t changing in this round of updates. I would be surprised if Google didn’t soon make some changes there, too (or unify its storage pricing across these services).
BigQuery, Google’s database for doing big data analysis, is getting the largest price drop at 85 percent. The team reduced per-gigabyte storage pricing from $0.08/GB to $0.026/GB, a 68 percent drop, and interactive queries now cost $5/TB instead of $35/TB. Batch queries now also cost $5/TB instead of the previous $20/TB.
As Google’s Direct of Product for the Cloud Platform Greg DeMichillie told me last week, the industry has somehow moved away from the original dream of the cloud over the last few years. Provisioning servers in the cloud was supposed to be simple and cheaper than doing it yourself. But setting up a cloud server now often involves plenty of work, and cloud pricing hasn’t exactly followed Moore’s law.
Public cloud prices have dropped 6 to 8 percent annually, but hardware costs actually dropped 30 to 45 percent annually. “It’s certainly true that there is more to the cloud than just hardware,” DeMichellie told me. “We are trying to rest this industry’s prices back to where they should’ve been if they followed this law.”
In an effort to simplify pricing, Google will now offer automatic discounts to users who run sustained workloads on its cloud. Instead of having to deal with reserved and pre-paid instances, which are their own kind of lock-in, Google will now automatically give all developers on the platform discounts when they use a machine over a longer period of time. These discounts kick in automatically after the first quarter of the month and get deeper as the month continues. Then prices reset again at the beginning of the month and the cycle repeats. Even if you just use the service for two separate weeks per month, the discount will kick in for the second week.
On Compute Engine, developers who use the service 24/7 can see price drops of over 50 percent. This way, if you want to change to a different instance type after a month, you aren’t locked in and you don’t lose your discounts.
Even without the sustained use discounts, Google’s pricing now undercuts that of its competitors for on-demand pricing and is often lower than Amazon’s EC2 prices for reserved instances, too.
So far, Google, Amazon and Microsoft have always matched one another’s cloud-hosting prices, and chances are, we will see price drops from Google’s competitors in the coming weeks, too. Amazon, of course, is hosting its own cloud-centric event later this week and we will likely hear more from them about pricing, too.
Google Announces Massive Price Drops For Its Cloud Computing Services And Storage, Introduces Sustained-Use Discounts
Última edição por 5ms; 25-03-2014 às 20:00.
25-03-2014, 21:01 #5
25-03-2014, 23:25 #6
- Data de Ingresso
- Oct 2010
- Rio de Janeiro
No free lunch...