據今日石油網2020年6月5日報道，開發能源需要一個大的工具箱，它需要依靠詳細的地質調查來發現石油和天然氣，需要依靠高性能的鉆井平臺來進入儲油層，最后，需要依靠可靠的運輸原油或天然氣的能力。美國石油巨頭?？松梨诠救涨氨硎?，開發能源還需要依靠強大的超級計算機和高速互聯網來成功運營他們的油氣田。

在開始生產之前，地球科學家，工程師和計算機科學家首先要建立一個石油或天然氣埋藏在地下深處儲集層的數字模型。如今，隨著越來越多的石油和天然氣儲藏在難以開采的地區，計算機演算法將提高安全運行和商業成功井的機會。

這些建立在大量數據之上的復雜模型需要花費大量時間來構建。但是?？松梨诠镜囊唤M科學家已經開發出了突破性的軟件，能夠在幾分鐘而不是幾天內開發出模型。

這個里程碑是在伊利諾伊大學的國家超級計算應用中心實現的，該中心擁有世界上最強大的被稱為“藍色水域”的超級計算機之一。除了能源，這個平臺還支持從物理學到火山研究等領域的先進研究。

埃克森美孚公司的研究項目涉及22000多臺計算機，總共擁有70多萬臺處理器。這足夠每秒進行1千萬億次計算。計算機的磁盤存儲空間大到足以存儲世界上所有圖書館的所有打印文件。

不過，這項成就并不僅僅在于原始的計算機能力，還在于巧妙地處理這些單獨的機器以有效地相互聯系并最終快速而準確地建立儲層模型的能力。

這一突破將加快油氣儲量的開發，有助于滿足世界對安全可靠能源日益增長的需求。

李峻 編譯自 今日石油

原文如下：

Supercomputers, high-speed internet key to operating oil fields

Developing energy requires a big toolbox. It relies on detailed geological surveys to find the oil and gas, high-performance drilling platforms to access the reservoirs and, finally, the ability to transport the crude or natural gas responsibly. US oil major ExxonMobil has said energy companies also depend on powerful supercomputers to successfully operate their oil and gas fields.

Before initiating production, geoscientists, engineers, and computer scientists first build out a digital replica of the reservoirs where the oil or gas is buried deep beneath the ground. These days, with more and more oil and gas reserves stored in hard-to-access areas, computer algorithms increase the chance of operating safe and commercially successful wells.

These complex models, built on reams of data, take a lot of time to build. But a group of ExxonMobil scientists has developed breakthrough software that is able to develop the models in minutes rather than days.

This milestone was achieved at the University of Illinois’ National Center for Supercomputing Applications, home to one of the world’s most powerful supercomputers, known as Blue Waters. In addition to energy, the platform supports advanced research in fields ranging from physics to the study of volcanoes.

The ExxonMobil research project involved more than 22,000 computers, which combined, held more than 700,000 processors. That’s enough computing power to perform 1 quadrillion calculations every second. The computer’s disk storage space is large enough to store all of the printed documents in all of the world’s libraries.

The achievement, though, isn’t just about raw computer power, but also the ability to finesse these individual machines to effectively communicate with one another and ultimately build the reservoir models quickly and accurately.

The breakthrough will speed up the development of oil and gas reserves that will help supply the world’s ever-expanding demand for safe and reliable energy.