Displaying items by tag: supercomputer
At Lenovo’s TRANSFORM event held in New York on June 20, the company unveiled an end-to-end data center portfolio that enables customers to harness the power of the “intelligence revolution” and create a strong technology foundation that supports transformative capabilities such as data analytics, high-performance computing, hybrid cloud, artificial intelligence and machine learning.
Lenovo introduced its largest, most comprehensive portfolio of server, storage, networking, software and data center services, as well as a new set of brands poised to take this leading position in customer satisfaction to even greater heights: ThinkSystem and ThinkAgile. The new portfolio of Lenovo solutions will be available in summer 2017.
“Today’s announcements mark a significant day in the next phase of Lenovo’s commitment to advancing the data center customer experience,” said Kirk Skaugen, President of Lenovo Data Center Group speaking at the event.
“Our leadership in x86 server customer satisfaction and x86 server reliability, along with these new ThinkSystem and ThinkAgile brands, and Lenovo’s new portfolio of data center solutions represent the most reliable, most agile, and highest performing data center solutions in the industry. We are confident we can empower our customers to solve their most challenging business problems and become digital disrupters in their own industries.”
This year, Lenovo celebrates 25 years of proven Think quality, performance and reliability, and Lenovo ThinkSystem and ThinkAgile data center solutions provide organizations with fluid resources that can adapt to the evolving needs of a business. According to a recent IDC survey of CIOs and technology leaders, 72 percent of respondents cited supporting digital transformation and business growth as the leading rationale in IT infrastructure decisions.
In the past three years, hybrid cloud significantly matured, while software defined architectures and hyperconverged infrastructure have disrupted the status quo for IT infrastructure in larger businesses, according to Lenovo. In response to these changes, customers face challenges including evolving their infrastructure strategies and considering how to best enable them now and for the future.
The new completely redesigned ThinkSystem portfolio by Lenovo, based on Intel® Xeon® Scalable processors, brings the gold-standard for the industry under a single unified brand spanning servers, storage and networking systems.
ThinkAgile is a completely new software-defined solutions portfolio built on the foundation of Lenovo ThinkSystem platforms: it adapts to changing IT needs while reducing complexity and cost created by silos in traditional IT. Even better together, the ThinkSystem and ThinkAgile portfolios empower customers to create the foundation of a future-defined data center with “simplicity, agility and flexibility, and future-proofed preparation.”
Often, customers find the price for technological advancement and advanced IT capabilities are complexity and confusion, according to Lenovo. The new ThinkSystem portfolio of servers, storage and networking solutions allows customers to address these concerns by giving them the ability to streamline IT infrastructure; this leads to increased service levels of data center operations that tie directly to business growth.
The ThinkSystem portfolio is engineered to work seamlessly with an organization’s previous investments, without the need to completely re-architect the data center. In addition, with the Lenovo ThinkAgile portfolio, integrated solutions adapt to changing IT and business needs by delivering the speed, simplified management and agility of cloud services while retaining the much-valued control and governance of on-premises IT.
Its pre-integrated, pre-built and pre-tested offerings deliver “robust new capabilities” to customers’ data centers, including automated lifecycle management, decreased TCO and reduced need for IT resources to maintain the platform. These solutions include the ThinkAgile SX series for Microsoft Azure Stack, as well as the ThinkAgile SX rack level solutions that integrate networking with hyperconverged offerings for a seamless customer experience.
Also, with the pace of technology innovation continuing to accelerate, customers will increasingly require the flexibility to adjust to changes and ensure the infrastructure they implement today stays relevant for their future needs. Lenovo designed the ThinkSystem and ThinkAgile solutions, such as the ThinkSystem SD530 high-performance computing offering, to support workloads spanning large data analytics applications to high performance computing, artificial intelligence and the largest hyperscale environments.
Lenovo is “committed to delivering transformational experiences for both its customers and the industry” and also announced the delivery of one of the most powerful systems – among the first in the world based on Intel Xeon Platinum processor family – to Barcelona Supercomputing Center (BSC).
The BSC MareNostrum 4 supercomputer, tasked with jobs in science and engineering research, incorporates 48 racks with more than 3,400 nodes with next generation Intel Xeon processors and a central memory of 390 Terabytes. Its peak power will be over 11 Petaflop/s, allowing the center to test and analyze the performance of the most recent developments in the field of supercomputing.
The newly launched x86 ThinkSystem platform provides us with ten times more processing power than the previous MareNostrum supercomputer, allowing BSC to continue the type of scientific research that enhances humanity's efforts to seek new knowledge and drive progress. Our experience with Lenovo has been easy and tightly integrated to overcome our challenges and achieve success, and we look forward to future collaboration," said Sergi Girona, Operations Director, BSC.
China is reportedly working towards developing a prototype exascale computer, which refers to computing systems capable of at least one exaFLOPS, or a billion, billion calculations per second. Such capacity represents a thousandfold increase over the first petascale computer that came into operation in 2008. If China is successful, it would secure the nation’s status as a leading power in supercomputing.
In June last year China made waves when it announced the completion of the world’s current fastest computer, the Sunway TaihuLight, which was twice as fast as the previous fastest model. The Sunway TaihuLight was a breakthrough for China, because it was made using only locally made microchips, making it the first time a country has made a top computer without using US technology.
But that computer pales in comparison to exascale computers which, if completed, could execute at least one quintillion (a billion, billion) calculations per second. A prototype of the supercomputer is currently in the works, according to Zhang Ting, an application engineer at the National Supercomputer Center in the port city of Tianjin, who spoke to Xinhua News agency, but a complete version of the prototype is still a few years off, he said.
“A complete computing system of the exascale supercomputer and its applications can only be expected in 2020, and will be 200 times more powerful than the country’s first petaflop computer Tianhe-1, recognized as the world’s fastest in 2010,” said Ting.
He added that the exascale computer could have applications in big data and cloud computing work, and that the prototype in China would lead the world in data transmission efficiency as well as calculation speed. China is flexing its technical muscle, last year having more top-ranked supercomputers than the U.S. featured on tracking website Top500.org, with 167 compared to 165.
Two of the world’s top ten fastest computers reside in China, and only five reside in the United States, as of November, according to the Top500 ranking. The other supercomputers reside in Japan and Switzerland. China is said to be pouring money into major science and technology projects to secure its leading position. However, despite some gains the country's scientific output still lags behind, and its universities generally fare poorly in global rankings.
China has created what Top500 considers to be the world’s most powerful supercomputer, installed at the National Supercomputing Centre in Wuxi, Jiangsu province. The 93 petaflop Sunway TaihuLight, at its peak, can perform around 93,000 trillion calculations per second. It marks the first time since the Top500 supercomputer list began that China has overtaken the U.S. with 167 computers on the list, while the U.S. has only featured 165 times.
The Top500 list, which announced the world’s top supercomputers on June 20, says the Sunway TaihuLight supercomputer is able to perform at a level three times as efficient as its predecessor Tianhe-2, also of Chinese make.
Jack Dongarra who wrote a paper about the machine, says the supercomputer’s main applications include advanced manufacturing, weather forecasting and big data analytics. It has more than 10.5 million locally-made processing cores and 40,960 nodes, running on a Linux-based operating system.
The Top500 says: “Considering that just 10 years ago, China claimed a mere 28 systems on the list, with none ranked in the top 30, the nation has come further and faster than any other country in the history of supercomputing.”
The United States has four supercomputers in the top 10 of the Top500 supercomputer list, while China has two which take up the top two spots. Other positions in the top 10 are occupied by machines from Germany, Switzerland, Japan and Saudi Arabia.
“As a computer scientist, it’s difficult writing software that can take advantage of and control large numbers of computer cores,” says Professor Les Carr from the University of Southampton. “This is why supercomputers are restricted to specialized applications – you need very specialized computing needs to take advantage of them.”