譯科技 | 引爆2020年的企業(yè)數據庫趨勢
來源：Dataversity 時間：2020-02-13 15:01:45 作者：Ariff Kassam
The 2020 Tipping Point： Enterprise Database Trends
數據觀 | 石煜倩（譯）
We’re at a unique inflection point in the database industry. Many things have changed over the last decade， but the cloud， containers， microservices， and the broader drivers for innovation – such as digital transformation – have created significant demand for new database technologies.
我們正處于數據庫行業(yè)的特殊拐點。在過去十年中，萬事萬物都發(fā)生了翻天覆地的變化，云業(yè)務、計算機容器、微服務以及更廣泛的創(chuàng)新驅動力（比如數字化轉型）對新興數據庫技術提出了更高要求。
While emerging social or web-based enterprises already have nearly all of their applications in the cloud， such applications represent a small portion of the existing database workload in traditional enterprises. These organizations have significant technical debt and thus only a small portion of their applications have been moved.
雖然新興社交或基于互聯網的企業(yè)幾乎已經將所有的應用程序都放在了云端，但這些應用程序只占傳統(tǒng)企業(yè)現有數據庫工作負載的一小部分，這些公司的技術還不足以支撐全部數據的轉移，因此僅有一小部分應用程序被轉移了。
And yet， enterprises are under intense pressure to modernize IT infrastructure in order to address key business challenges and respond quickly to emerging opportunities. So how can they find a low-risk way to adopt new technology and be as agile as their web-based counterparts？ Here are three ideas：
然而，企業(yè)目前面臨著巨大的壓力，因為他們需要對IT基礎架構進行現代化改革以應對主要的業(yè)務挑戰(zhàn)，并迅速對新出現的機遇做出反應。那么，他們如何才能找到一種低風險的方法來采用新工藝，并像基于互聯網的同行一樣敏捷呢？下面將介紹三種想法：
1. Capitalize on cloud and container-native applications利用云原生①與容器原生應用
（注釋①：云原生技術有利于各組織在公有云、私有云和混合云等新型動態(tài)環(huán)境中，構建和運行可彈性擴展的應用。云原生的代表技術包括容器、服務網格、微服務、不可變基礎設施和聲明式API。）
Cloud providers will continue to compete with on-premises offerings based on increased services and the potential to lower costs. However， your cloud provider of choice today should not prevent you from switching to a different cloud in the future. Avoiding cloud “l(fā)ock in” is not a new concept， but it has taken on new urgency as more and more critical business applications and heavy transactional systems move to the cloud. By default， software applications that are cloud and container-native – built and architected to perform in a cloud or container environment – will garner more traction.
云提供商將繼續(xù)與服務不斷增加、成本不斷降低的本地產品競爭。但是，現有云服務商不應該阻止用戶未來切換到不同的云端。避免云“鎖定”并不是一個新概念，但是它呈現出了新的緊迫性，因為越來越多的關鍵業(yè)務應用和繁重的事務性系統(tǒng)被遷移到云端。默認情況下，云原生與容器原生軟件應用（在云或者容器環(huán)境下構建和運行）對用戶的吸引力會更大。
These applications are designed for scalability， resiliency， and flexibility， enabling the infrastructure to keep pace with business demands. This goes beyond a simple containerization approach. Container-native applications are optimized to take advantage of all the benefits that containers and the cloud have to offer. For example， organizations can speed the application delivery process by capitalizing on container runtime management and orchestration platforms such as Kubernetes.
這些應用程序旨在實現可伸縮性、彈性和靈活性，使基礎設施能夠滿足業(yè)務需求。這超出了簡單的容器化解決方法。容器原生應用經過優(yōu)化，可以利用容器和云提供的所有優(yōu)勢。例如，公司可以利用容器運行時②管理和編排平臺（如Kubernetes）來加速應用程序交付過程。
（注釋②：容器運行時指的是能夠基于在線獲取的鏡像來創(chuàng)建和運行容器的程序。）
2. Leverage Kubernetes to reduce complexity and costs利用Kubernetes③來降低復雜性和成本
As software developers are pushed to speed up the process from application development to production， many have discovered that building and deploying stateless applications in containers is relatively easy. These types of applications can quickly scale out with increased user demands and protect against failures. There’s just one catch： The vast majority of applications used by enterprises （personalized applications or those that handle business-critical data） require a persistent state. They require storing data to protect against failures so that the application won’t lose its state. As it turns out， these stateful applications are harder to fit into the new world of microservices and containers.
（注釋③：kubernetes，簡稱K8s，是用8代替8個字符“ubernete”而成的縮寫。是一個開源的，用于管理云平臺中多個主機上的容器化的應用。）
隨著軟件開發(fā)人員不斷加快從應用程序開發(fā)到生產的過程，許多人發(fā)現在容器中構建和部署無狀態(tài)應用程序相對容易。這些類型的應用程序可以隨著用戶需求的增加而快速擴展，并能夠防止出現故障。只有一個問題：企業(yè)使用的絕大多數應用程序（個性化應用程序或處理關鍵業(yè)務數據的應用程序）都需要保持持續(xù)狀態(tài)。它們需要存儲數據以防止出現故障，這樣桌面內容才會保持一致。事實證明，這些有狀態(tài)應用程序很難適應微服務和容器。
Kubernetes has emerged as the de facto way to simplify and automate control of container processes. Databases have long been the standard for managing state for applications. To work in containers and the Kubernetes orchestration environment， databases must be able to support the dynamic nature of Kubernetes. Most traditional databases assume a static environment and do not work well within Kubernetes. As enterprises move more critical applications to the cloud， automating and simplifying deployment of distributed， stateful applications in Kubernetes will only increase.
Kubernetes已經成為簡化和自動控制容器過程的實際方法。長期以來，數據庫一直是管理應用程序狀態(tài)的標準。要在容器和Kubernetes編排環(huán)境中工作，數據庫必須能夠支持Kubernetes的動態(tài)性。大多數傳統(tǒng)數據庫采用靜態(tài)環(huán)境，在Kubernetes中不能很好地工作。隨著企業(yè)將更多主要的應用程序遷移到云中，在Kubernetes中自動化和簡化分布式、有狀態(tài)應用程序的部署只會增加。
3.Accelerate a multi-cloud strategy加快實施“多云戰(zhàn)略”
In an always-on world， availability is a key requirement for application success. And yet high-profile outages from major cloud vendors persist. To combat that risk of data loss and potential customer impact during such instances， organizations can accelerate the adoption of multi-cloud， or even better， inter-cloud strategies. This is particularly relevant in heavily regulated industries such as banking. To run core banking applications in the cloud – such as mortgage applications， credit card processing， and even fraud prevention – supporting multi-cloud environments becomes critical to provide the necessary level of data protection and availability.
在一個始終在線的世界中，實現可用性是應用程序成功的關鍵。然而，來自主要云提供商的停機問題仍然持續(xù)存在。為了在這種情況下避免數據丟失和潛在客戶的影響，相關組織機構可以加速采用多云甚至跨云策略。這在銀行業(yè)等監(jiān)管嚴格的行業(yè)尤為重要。要在云中運行核心銀行應用程序（如抵押貸款應用程序、信用卡處理、甚至欺詐預防），支持多云環(huán)境對于提供必要的數據保護和可用性至關重要。
In the future， specific banking regulations may require deployment in more than a single cloud （inter-cloud）. As the ability to deploy stateful applications reliably grows， there will be greater emphasis on the need to provide continuous availability through the use of distributed deployments and multiple clouds.
未來，特定的銀行管理制度可能需要在多個云（云間）中進行部署。隨著部署有狀態(tài)應用程序的能力不斷增強，通過使用分布式部署和多個云來提供持續(xù)可用性的需求將會被進一步強調。
As organizations continue down this path of application modernization and business agility， traditional enterprises can develop the modern architecture needed to compete in today’s on-demand world. By capitalizing on the opportunities to move to the cloud or distributed environments – leveraging technologies that are cloud or container-native and managing application complexity with new tools – these organizations can ensure they are ready to meet customer demands in 2020 and beyond.
隨著各組織機構將繼續(xù)朝著應用程序現代化和業(yè)務敏捷性的道路發(fā)展，傳統(tǒng)企業(yè)可以開發(fā)在當今競爭世界中所需的現代體系結構。通過利用遷移到云或分布式環(huán)境的機會——利用云原生或容器原生技術并使用新工具管理應用程序的復雜性——這些組織可以確保它們能夠滿足2020年及以后的客戶需求。(石煜倩)
?
注：《譯科技|引爆2020年的企業(yè)數據庫趨勢》來源于dataversity網站（點擊查看原文）。本文系數據觀原創(chuàng)編譯，譯者數據觀/石煜倩，轉載請務必注明譯者和來源。
責任編輯：張薇