Intel 推出第 3 代 Xeon Scalable 處理器加入 Crypto 加速與 DL Boost 帶給資料中心強大效能

]

從資料中心、雲端、網路到 AI,Intel 推出第 3 代 Xeon Scalable 處理器(“Ice Lake”),採用 10nm 製程打造最高 40 核心,並有著 8 通道DDR4-3200 記憶體支援性,以及 64 lane PCIe Gen4 的擴充能力。而第 3 代 Intel Xeon 可擴充處理器,比起上一代效能有著顯著提升,於熱門資料中心工作負載提供平均 46% 提升改善。處理器還增加新款與強化後的平台功能,包含內建安全性 Intel SGX ,以及 Intel Crypto Acceleration 和針對 AI 加速的 Intel DL Boost。

這些結合 Intel Select Solutions 和 Intel Market Ready Solutions 等英特爾廣泛產品組合的新功能,讓客戶能夠加速佈署橫跨雲端、AI、企業、HPC、網路、安全性與邊緣的各種應用。

內建 AI 加速:最新第 3 代 Intel Xeon 可擴充處理器提供 AI 效能、生產力、簡易性,讓客戶能夠洞悉他們的資料並解放更多具有價值的內容。相較前一世代可提升 74% 快的 AI 效能;相較 AMD EPYC 7763,於廣泛應用的 20 項 AI 混合工作負載,最高提供 1.5 倍的效能:相較於Nvidia A100 GPU,於廣泛應用的 20 項 AI 混合工作負載,最高提供 1.3 倍的效能。

內建安全性:經過成千上萬的考察研究與生產佈署,加上隨著時間不斷淬鍊,Intel SGX 以系統內最小潛在攻擊面保護敏感程式碼與資料。如今可於雙插槽 Xeon 可擴充處理器的指定位址空間(enclave)使用,該指定位址空間能夠隔離與處理最高 1TB 的程式碼與資料,以便支援主流工作負載的需求。結合 Intel Total Memory Encryption 和 Intel Platform Firmware Resilience 在內的新功能。

內建加密加速:Intel®Crypto Acceleration 於許多重要的加密演算法提供突破性的性能。執行密集加密的商業模式能夠從此功能當中獲益,如每天處理數以百萬計消費交易的線上銷售,於保護消費者資料的同時,無需影響使用者回應時間或是整體系統效能。

此外,為加速第 3 代 Intel Xeon 可擴充平台的工作負載,軟體開發者能夠透過開放、跨架構程式設計的 oneAPI 最佳化他們的應用,從專利模式的技術和經濟負擔當中解放。Intel oneAPI 開發工具包透過進階編譯器、函式庫、分析與除錯工具,實現處理器的AI與加密功能和效能 。

領先業界的資料中心平台

市場最為普及的英特爾資料中心平台,具備無與倫比的移動、儲存與處理資料的能力。最新第 3 代Intel Xeon 可擴充平台包含 Intel Optane Persistent Memory 200 系列、Intel Optane Solid State Drive(SSD)P5800X 和 Intel SSD D5-P5316 NAND SSD,以及 Intel Ethernet 800 系列網路介面卡和最新的 Intel Agilex FPGA。上述可在第 3 代 Intel Xeon 可擴充平台產品資訊當中獲得更多資訊。

橫跨雲端、網路與智慧邊緣提供靈活效能

最新第 3 代Intel Xeon 可擴充平台為廣大範圍的市場區隔最佳化-從雲端到智慧邊緣。

為雲端而生:第 3 代 Intel Xeon 可擴充處理器為滿足雲端工作負載和支援大範圍服務環境的苛刻需求而打造並最佳化。全球超過 800 個雲端服務供應商於 Intel Xeon 可擴充處理器上執行,且所有最大的雲端服務供應商正計畫於 2021 年,由第 3 代 Intel Xeon 可擴充處理器支援其雲端服務。

為網路而生:英特爾的網路最佳化「N 版本」專為支援多樣的網路環境而設計,並為多種工作負載和性能水準最佳化。最新第 3 代 Intel Xeon 可擴充處理器相較前一世代,於一系列廣泛佈署的網路與 5G 工作負載,效能平均提升 62%。與超過 400 位 Intel Network Builders 成員的廣泛生態系一同工作,英特爾提供基於第 3 代 Intel Xeon 可擴充處理器「N 版本」的解決方案藍圖,進而加快 vRAN、NFVI、虛擬 CDN 等的測試認證並縮短佈署時間。

為智慧邊緣而生:第 3 代 Intel Xeon 可擴充處理器針對強而有力的 AI、複雜的影像或視訊分析,以及智慧邊緣整合工作負載,提供所需的效能、安全性和作業控制。相對於前一世代,平台為影像分類提供最高1.56倍的AI推論效能。

Could a ‘Crypto Climate Accord’ erase cryptocurrencies’ carbon footprint?

]

A newly announced “Crypto Climate Accord” aims to erase cryptocurrencies’ legacy of climate pollution. That’s a tall order considering the enormous amounts of energy that the most popular cryptocurrencies — bitcoin and Ethereum — consume. The loose goals laid out in the plan so far face potentially insurmountable challenges.

A 2040 target for the crypto industry to reach “net zero” emissions

The “accord” is led by the private sector — not governments — and outlines a few preliminary objectives. It seeks to transition all blockchains to renewable energy by 2030 or sooner. It sets a 2040 target for the crypto industry to reach “net zero” emissions, which would involve reducing pollution and turning to strategies that might be able to suck the industry’s historical carbon dioxide emissions out of the atmosphere.

Lastly and perhaps most realistically, it aims to develop an open-source accounting standard that can be used to consistently measure emissions generated by the crypto industry. They also want to develop software that can verify how much renewable energy a blockchain uses.

If achieved, those goals would solve a very real problem. Bitcoin alone has roughly the same carbon footprint annually as Hong Kong, while Ethereum’s annual carbon emissions rival Lithuania’s. Their climate pollution is growing even as scientists’ research warns that global emissions need to be cut almost in half this decade to avoid the worst effects of climate change.

The accord has support from some influential names in climate action and the crypto industry — including cryptocurrency company Ripple, blockchain technology conglomerate Consensys, billionaire climate crusader Tom Steyer, and the United Nations-appointed “climate champions.”

While tackling the environmental damage done by the crypto industry might be a worthy challenge, critics say the broad goals are unlikely to result in meaningful change.

“Some things just can’t be fixed,” says economist Alex de Vries.

“Some things just can’t be fixed.”

Unfortunately for the Crypto Climate Accord, bitcoin is the biggest player in the game, and it’s likely to cause the accord the most trouble because of how much energy it uses. Bitcoin is purposely inefficient — which is a problem renewables can’t fix. It uses a model called “proof of work” to keep its ledgers secure. “Miners” who verify transactions to get new coins do so by using energy-guzzling machines to solve increasingly difficult puzzles. (Ethereum also uses proof of work but has said for years that it will eventually transition to another model.)

Related The climate controversy swirling around NFTs

Those machines will continue to compete for renewable energy with arguably more essential needs, like keeping the power on in people’s homes. And if cryptocurrencies increase electricity demand beyond available renewable resources, utilities might turn to fossil fuels. That’s why cleaning up energy sources and increasing energy efficiency are two sides of the same coin when it comes to tackling climate change.

Regardless, the accord’s founders are optimistic about a greener future for bitcoin. “I’ve been in conversation with folks from the bitcoin ecosystem, it’s a pretty simple pitch,” says Jesse Morris, chief commercial officer of the nonprofit Energy Web Foundation, which is spearheading the initiative. “If we can make bitcoin green, it will be much easier and lower risk for other organizations to come in and buy more Bitcoin.”

“If we can make bitcoin green, it will be much easier and lower risk for other organizations to come in and buy more bitcoin.”

Bitcoin still accounts for more than half of the entire cryptocurrency market capitalization. But it is facing competition from newer cryptocurrencies that have found ways to be greener. Other cryptocurrencies use different blockchain technology than bitcoin and consume very little energy in comparison as a result. For those cryptocurrencies, like Ripple’s XRP, running on renewables could be more feasible.

And while renewable energy costs have fallen dramatically, luring bitcoin miners to places with abundant renewable energy would likely require heavy subsidies to keep them from turning to cheaper, dirtier fuel sources, de Vries says. “Just the sound of that — It sounds really wrong,” he says. “Why would you want to subsidize an industry that uses energy just because it is set up to waste resources?”

The new crypto accord, however, is “not about coming together and asking for subsidy by any means,” says Morris. “We just want to get everybody together and start nailing the action here.” Many blockchains, like bitcoin, were designed to be a decentralized system with no top-down oversight. So getting everyone on board, even within a single blockchain, will be a huge task.

The accord’s objectives are supposed to be fleshed out and finalized by the time a big United Nations climate conference rolls around in November. But Morris admits that the initial plans chase big aspirations rather than fine details. “So many of these other decarbonization efforts are very much thinking their way into acting,” Morris says. “Whereas in Crypto, because it’s kind of the Wild West, it’s about acting our way into new thinking.”

Crypto-Conservation

]

By using blockchain-based chips to tag endangered fish, most notably the critically endangered Apache trout, the Xenia Project has been able to gather the necessary data to confirm the root cause of this fish’s decline, information necessary to take appropriate measures to combat their extinction.

Wilk moved from New Jersey to Arizona with his parents when he was five. He has always enjoyed the outdoors, and is especially interested in fishing. A longtime Boy Scout, he has gone on many outings over the years, fishing with his buddies in area rivers. Early on, they used unconventional tactics like tying Slim Jim pieces to lines of string, something he says was surprisingly effective for catching crayfish. Later, as their casting skills improved, they progressed to rods and reels and aimed for larger catch like bass and pike.

Another formative experience for him, his first glimpse of the ocean at the age of nine, happened during a trip to visit his aunt in Galveston, Texas. The year was 2010, and his visit coincided with the Deepwater Horizon oil spill off the coast of Louisiana . Although Wilk knew about the spill before he arrived and had seen some photos, “to see it in real life like that was traumatizing,” he says. “To be met with black sludge, birds, and fish floundering on the rocks was pretty shocking. It really stuck with me through the years.”

Back in Arizona, a state well known for its waterless tracts of dirt, he began to notice a common misconception: The idea that aquatic life there is not a concern. Wilk knew of several species in local rivers that were in jeopardy, including the Gila chub (endangered) and the Apache trout (critically endangered). The reasons for their declining population wasn’t known, but Wilk believed blockchain technology could help uncover the answers.

He came up with the idea of designing a blockchain-based microchip that could be attached to endangered fish as a tracking device for gathering data during migration. After developing the idea fully, he reached out to area universities hoping to confer with technology experts, but the response was predominantly skepticism. Fortuitously, Paul L. Bailey, a professor of math and computer science from Southern Arkansas University, was filling in for Wilk’s advanced-level math class. The course was linear algebra and vector calculus, and after Wilk approached Bailey, he took an interest in Wilk’s ideas. Bailey helped Wilk with the actual technological structure of the chip and also helped him refine his documentation. Bailey passed away earlier this year, but looking back, Wilk says that the late professor’s willingness to share his time was “a testament to his character.” He also credits Bailey with helping him to outline his ideas in a way that would encourage environmental groups and other experts to be more receptive.

One organization, the Black River Conservation Society (BRCS), was the first to agree to collaborate. This small sustainability group focuses on the conservation of aquatic species in Arizona’s Black River (which runs through Arizona’s White Mountains and is the native environment of the Apache trout).

“Ethan reached out to me almost two years ago to propose a partnership,” recalls Noah Williamson, who was president of BRCS at the time, in an email exchange. Wilks offered to provide the BRCS with his blockchain-based chips if they would supply him with volunteers and other resources to support him in doing the actual tagging. “To tell you the truth,” confesses Williamson, “I didn’t have much faith in his tech at first, but it proved its worth, and we continued to use it to tag trout in the area over the next year or so.”

Wilk had also been holding monthly fundraisers to raise money and local awareness of marine sustainability issues. He often hosted dodgeball games, using half-time to explain what the Xenia Project was all about. In addition to the donations he received, he gathered volunteers — hundreds of them, including 40 of his fellow students. Soon a larger like-minded organization joined in to support him: the Arizona Sustainability Alliance, an organization focused on supporting cutting-edge technology for environmentally sustainable projects.

The tagging began in early 2017 under the guidance of local forestry agencies. All volunteers received about an hour of training in how to tag the fish properly prior to heading into the field. The microchips are about the size of a fingertip and are attached to a zip tie, which is fastened either around the trout’s fins or through its gills, depending on which method causes the least harm to the species.