讯
全球新闻情报终端
智能情报台
在线
已显示 2/10
近7天
实时新闻情报台

新闻

英文原文版:支持全部新闻、关键词检索、时间范围、个性化推荐和下载。
当前新闻
2
当前可见新闻
关键词模式
microsoft
关键词筛选
排序方式
最新时间
当前排序规则
当前视图
当前显示 2 / 10 条,关键词:microsoft,范围: 近7天, 排序:最新时间
数据由系统缓存并自动更新 下载结构化数据 下载表格 下载文档
预览访问
当前为游客浏览模式,仅展示前 2 条新闻。 登录后可解锁全部内容和下载功能。 去登录
来源 Flexsin Technologies
发布时间
世界协调时 2026-10-02 11:30
北京时间 2026-10-02 19:30
作者 Munesh Singh 地点 United States、France
A growing share of government agencies, healthcare providers, and financial institutions moving to the cloud are doing so under one specific condition: the data can't leave a defined jurisdiction, and no outside party including the cloud provider itself gets to touch it without authorization. That's a sovereign cloud. And it changes the migration playbook in ways a standard lift-and-shift never has to account for. Take a manufacturing company holding export-controlled engineering drawings. It's
展开查看正文
A growing share of government agencies, healthcare providers, and financial institutions moving to the cloud are doing so under one specific condition: the data can't leave a defined jurisdiction, and no outside party including the cloud provider itself gets to touch it without authorization. That's a sovereign cloud. And it changes the migration playbook in ways a standard lift-and-shift never has to account for. Take a manufacturing company holding export-controlled engineering drawings. It's not enough to ask whether the workload can run in the cloud. Someone has to know exactly who can technically access the files, where the bytes physically sit, and what happens if a provider is legally ordered to hand them over to a foreign government. A cloud migration company handling sovereign work answers those three questions before a single VM gets provisioned. What Sovereign Cloud Migration Actually Means Sovereign cloud is infrastructure operated and frequently owned within the same legal jurisdiction as the data it holds. Region selection alone doesn't get you there. A customer can pick a data center in France and still be using a provider incorporated in the United States, which means U.S. laws like the CLOUD Act can compel disclosure no matter where the servers sit. Closing that gap takes more than a checkbox. Infrastructure providers need local legal entities. Encryption keys need to be placed with the customer, not the provider. Support staff with any access to the environment needs to be based inside the same jurisdiction as the data. A handful of providers package this as a dedicated product tier. Most customers end up building custom architecture on top of standard cloud services instead. Why Sensitive Workloads Need a Different Migration Approach A marketing website and a national identity database both technically run on virtual machines and object storage. Beyond that, the similarities end. Legal exposure on the sensitive side reshapes almost every decision in the plan. Sovereign workloads need network isolation from everything else dedicated to virtual private clouds, no shared routing that could leak traffic outside the jurisdiction. Access management needs logging granular enough to reconstruct, after the fact, exactly who touched which record and when. Disaster recovery sites must stay inside the same legal boundary as production, which rules out a surprising number of default DR configurations for cloud platforms set up on their own. Testing gets an extra layer, too. It's not enough to confirm that the application still works after the move. Someone must verify that no replication job, monitoring agent, or third-party integration is quietly shipping telemetry to a server outside the approved boundary. This is where migrations quietly fail a CDN or logging tool, left on its default settings, routes traffic through a country nobody approved. Data Residency: Where Your Data Physically Lives Data residency looks straightforward on a slide. In practice, a single transaction might touch a database in one region, a caching layer in another, and a monitoring dashboard run by a vendor whose servers sit somewhere else entirely. Getting an accurate picture means inventorying every service that touches sensitive data databases, message queues, caching layers, log aggregators, backup targets, and any SaaS tool with API access. Each one gets checked individually. A compliant primary database doesn't mean the rest of the stack is compliant, too. PostgreSQL deployments, for instance, often need more than a region setting adjusted. Replication targets, backup destinations, and even query logs can quietly push data across a boundary if left at defaults. Object storage buckets need region locking explicitly turned on, since a single misconfigured replication policy is enough to move data somewhere it shouldn't be. Access Control and Key Management Residency answers where the data sits. Access control answers who can actually read it and for sovereign workloads; that list needs to exclude the cloud provider's own engineers by default. Customer-managed encryption keys handle most of this. Sometimes called bring-your-own-key, the model has the provider storing encrypted data without ever holding an unsupervised decryption key. A few sovereign offerings push further into confidential computing, encrypting data even while it's being processed in memory rather than only at rest or in transit. Role-based access needs tight scoping. Break-glass procedures for emergency access should trigger automatic alerts and demand documented justification, not just a password reset. APIs linking the sovereign environment to anything else should run on short-lived tokens instead of static credentials as a leaked static key is a much longer-lived problem. Compliance Frameworks That Shape Sovereign Migrations Every sector and region layer on its own rules, and the migration plan has to be built against whichever framework actually applies. GDPR sets the baseline for data processing and cross-border transfer in the EU. HIPAA adds constraints for healthcare data in the U.S. India's Digital Personal Data Protection Act does something similar for a different jurisdiction entirely. Government workloads often need certification against FedRAMP or an equivalent national scheme, and those frameworks get specific encryption standards, logging depth, even personnel vetting requirements. Skip mapping these early, and expensive rework tends to show up later, usually right after an audit finds the gap. Choosing the Right Cloud Migration Company Plenty of vendors are comfortable with a standard AWS or Azure migration. Fewer have configured something like AWS European Sovereign Cloud correctly, and the difference shows up in the details nobody notices until it's a problem. Ask a prospective provider how they handle data mapping during discovery. Ask what their key management approach actually looks like, not just whether they "support" customer keys. Ask if they've run compliance audits alongside similar migrations before and get specifics. A provider treating sovereignty as a box to tick rather than an architectural principle tends to miss the configuration detail that only surfaces during an audit or a breach. Frequently Asked Questions: What makes a cloud sovereign versus just regional? Regional cloud stores data in a specific location. Sovereign cloud goes further, adding legal and operational controls that keep the provider's own staff and jurisdiction from accessing or being compelled to release the data. Can existing infrastructure be converted to sovereign cloud, or does it need a rebuild? Depends on the provider. Some offer sovereign tiers that existing workloads can shift into configuration changes. Others require rebuilding entirely separate infrastructure. How long does a sovereign cloud migration typically take? Timelines vary by workload, but discovery and compliance mapping alone usually run several weeks longer than they would for a standard migration. Does sovereign cloud cost more than standard cloud migration services? Usually, yes dedicated infrastructure, localized staffing, and compliance certification all add cost, though the premium shifts by provider and region. Is encryption alone enough to meet sovereignty requirements? No. Encryption covers confidentiality. Sovereignty also requires jurisdictional control over key access, personnel, and where the physical infrastructure sits. A Practical Migration Path The path usually runs through five stages: discovery, architecture design, pilot migration, full migration, and validation. Discovery is the full data and service inventory described earlier. Architecture design maps each workload to compliant infrastructure and locks in encryption and access policy. The pilot typically one non-critical workload tests whether the sovereignty controls hold before the rest of the environment follows. Full migration then moves in phases, with validation at each stage confirming residency and access boundaries hold under real traffic, not just in a configuration file. People Also Search For: 1. What is data sovereignty in cloud computing?The principle that data is subject to the laws of the country where it's collected or stored, which determines who can legally access or process it. 2. Which cloud providers offer sovereign cloud options?AWS, Microsoft Azure, and Google Cloud all offer sovereign or region-specific tiers, typically through local partner entities to meet jurisdictional requirements. 3. What industries require sovereign cloud migration?Government, healthcare, financial services, and defense most often require it, driven by strict data protection and national security rules. 4. How is sovereign cloud different from a private cloud?Private cloud is about dedicated infrastructure for one customer. Sovereign cloud is specifically about jurisdictional control over the data itself.
来源 Yahoo! Finance
发布时间
世界协调时 2026-10-02 11:20
北京时间 2026-10-02 19:20
作者 Motley Fool、Daniel Foelber 地点 New York
On Sept. 28, Nvidia (NASDAQ: NVDA) authorized a $150 billion increase to its share repurchase program, bringing the total to $235 billion. The new program is larger than Apple's (NASDAQ: AAPL) $100 billion additional authorization -- announced on April 30. Nvidia is now on track to surpass Apple in buybacks and dividends, a prediction I made in mid-September. Here's what the buyback increase means for Nvidia investors, why Nvidia and Apple's investment thesis have a core similarity, and which g
展开查看正文
On Sept. 28, Nvidia (NASDAQ: NVDA) authorized a $150 billion increase to its share repurchase program, bringing the total to $235 billion. The new program is larger than Apple's (NASDAQ: AAPL) $100 billion additional authorization -- announced on April 30. Nvidia is now on track to surpass Apple in buybacks and dividends, a prediction I made in mid-September. Here's what the buyback increase means for Nvidia investors, why Nvidia and Apple's investment thesis have a core similarity, and which growth stock is the better buy now. Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same "Total Conviction" signal is flashing for a company 1/100th the size of Nvidia. Continue " AI is contributing to Nvidia and Apple's cash flow Nvidia is providing the chips and networking needed to fuel increased AI compute demand. In contrast, Apple's devices and services provide how many consumers, students, and enterprises will interact with AI tools. Because of their positioning along the AI value chain, both companies benefit immensely from AI spending. By comparison, hyperscalers like Alphabet, Microsoft, and Amazon are draining their FCF as they spend heavily on AI. Anthropic and OpenAI depend on Apple as the touchpoint by which many users interact with their frontier models. As AI use cases and overall compute demand increase, Nvidia and Apple won't be pressured to shell out record capital expenditures. Instead, they are well-positioned to convert AI-driven revenue into FCF, which fuels buybacks and dividends. Apple tends to have the largest buyback program among U.S. companies. In its latest quarter, which was the third quarter of fiscal 2026, it spent $25.9 billion on buybacks and $4 billion on dividends -- ahead of Nvidia's $19.7 billion in buybacks and $6 billion in dividends in its latest quarter. But Nvidia is growing its capital return program much faster than Apple. A year ago, in the second quarter of fiscal 2026, Nvidia's buybacks and dividends were a combined $10 billion. Over half of Nvidia's free cash flow is going toward buybacks and dividends Nvidia has a clearer path to growing its FCF than Apple. For fiscal 2028, Nvidia is already guiding for a 70% increase in revenue. And its margins should remain sky high thanks to insatiable demand for the Vera Rubin compute platform. On its August earnings call, Nvidia said that it returned 60% of first-half fiscal 2027 FCF to shareholders, ahead of its 50% target. But going forward, it intends to increase the return of FCF net of strategic uses. Or, put another way, if Nvidia rakes in more FCF than it needs, the percentage of FCF returned to investors could increase.

还有 8 条新闻未解锁

登录后可解锁全部内容和下载功能。