Object Storage Service:OSS getting started FAQ

General FAQ

• What is Alibaba Cloud OSS?

  Alibaba Cloud Object Storage Service (OSS) is a secure, cost-effective, and highly durable cloud storage service with unlimited scalability. It is designed to provide at least 99.9999999999% (twelve 9s) of data durability and 99.995% of service availability.

• What can I do with OSS?

  OSS provides platform-agnostic RESTful APIs that allow you to store and access any type of data from any application, anytime, anywhere. Because OSS is highly scalable and you pay only for what you use, you can start small and scale as your business grows without compromising performance or data durability.

  You can use the APIs, SDKs, or migration tools provided by Alibaba Cloud to easily migrate large volumes of data into or out of OSS. After your data is stored in OSS, you can use the Standard storage class as the primary storage for mobile apps, large websites, image sharing, or popular audio and video content. For data that you access less frequently, you can use lower-cost storage classes with longer retention periods, such as Infrequent Access, Archive, Cold Archive, or Deep Cold Archive.

  For more information, see Overview of OSS features.

• Who should use OSS?

  OSS is ideal for website administrators of online communities, multimedia sharing sites, and e-commerce platforms, as well as for developers of mobile and software applications, game development companies, and any user with large-scale data storage needs.

  • Online audio, video, and image applications: Use the rich set of RESTful APIs provided by OSS to build massively distributed data storage solutions for use cases like short video storage, live stream recording, video on demand, photo-based social media, and video albums.

  • Education industry: Online education platforms, such as K-12 and online English learning services, can store data in OSS and combine it with OSS Transfer Acceleration or CDN to create solutions for massive data storage and content delivery.

  • AI and IoT: In scenarios such as autonomous driving and IoT, you can use Data Transport to collect and migrate training data to OSS. In home or public video surveillance scenarios, cameras can upload video data directly to OSS. This enables real-time video playback in apps and helps you reduce costs and meet compliance requirements by tiering data with lifecycle management.

  • Film and rendering industry: OSS provides highly elastic and massive storage for data such as production materials and media assets in film production and management. By integrating with Intelligent Media Management (IMM), you can create a solution that combines storage with intelligent data processing.

  • Genomics industry: OSS meets the data storage needs for genomics workflows, including gene sequencing, delivery, and diagnostics. Combined with powerful cloud computing capabilities, you can implement a comprehensive big data solution for storage, computation, and analysis.

• What kind of data can I store in OSS?

  OSS is suitable for storing attachments, high-definition images, audio and video files, and backups for forums and software applications. It can also store files for various apps, multi-device synchronization software, and file-sharing websites.

• What problems can developers solve with OSS that they cannot with on-premises solutions?

  OSS allows any developer to leverage the scale of Alibaba Cloud without upfront investment or compromising performance. It frees developers to focus on innovation rather than worrying about performance bottlenecks or security issues that come with business growth. OSS is not only cost-effective but also simple to use.

• How much data can I store in OSS? Is there a limit?

  OSS imposes no limit on the total storage capacity or the capacity of a single bucket. From the OSS console, you can upload objects smaller than 5 GB. To upload objects larger than 5 GB, you must use multipart upload or upload them directly by using ossbrowser 1.0 or the ossutil 2.0 command-line tool.

• What storage classes does OSS offer?

  OSS provides Standard, Infrequent Access, Archive, Cold Archive, and Deep Cold Archive storage classes to cover a full range of data storage scenarios, from hot to cold data. For more information, see Storage classes.

• How do I choose the right OSS storage class?

  The five OSS storage classes differ in billable size, minimum storage duration, restoration time, and data retrieval methods. You can choose different storage classes based on data access frequency and application scenarios to reduce storage costs.

  For example, if 70% of your data is not accessed for more than 30 days, this portion is considered colder data. We recommend storing this data in the Infrequent Access or Archive storage class to lower costs. You can also set lifecycle management rules for your data, and OSS will automatically transition colder data to the Infrequent Access or Archive storage class based on your rules. The colder the data, the lower the storage cost.

  Note that for data in the Archive or Cold Archive storage classes, data restoration takes minutes to hours, and real-time access is unavailable. OSS also charges an additional fee for data restoration.

• Does Alibaba Cloud use the data I store in OSS?

  Alibaba Cloud does not use or disclose your business data without authorization, except to fulfill your service requests or comply with legal and regulatory requirements. For more information, see the Product Terms of Service.

• Does Alibaba Cloud store its own data on OSS?

  Yes. Developers within Alibaba Cloud also use OSS as the authoritative data store for many projects. These projects rely on OSS for critical business operations.

• How does OSS ensure service availability if traffic from my application suddenly spikes?

  OSS was designed from the ground up to handle high traffic from any internet application. Its pay-as-you-go pricing and unlimited capacity ensure your service is not interrupted by traffic spikes. OSS automatically balances loads so that no single application is affected by traffic peaks.

• How is data organized in OSS?

  OSS provides key-value storage. When you store an object, you assign it a unique name, or key.

  Folders in OSS are a logical concept. When using an API or SDK, you can simulate a folder by including a directory path in the object's key. For example, if you define an object's key as dir/example.jpg, OSS creates a logical folder named dir in the current bucket and an object named example.jpg inside it. If you delete the object dir/example.jpg and the folder becomes empty, the logical folder dir will no longer exist.

• What intelligent features does OSS provide?

  OSS seamlessly integrates with numerous computing frameworks, including Hadoop, Spark, MaxCompute, BatchCompute, High-Performance Computing (HPC), and EMR. To reduce operational complexity, OSS offers easy-to-use SaaS services such as image processing and content moderation. Additionally, OSS provides Intelligent Media Management, which can quickly integrate various smart media processing algorithms, significantly improving the efficiency of media content management and distribution.

• How do I get started with OSS?

  1. Before you use Alibaba Cloud OSS, make sure you have registered an Alibaba Cloud account. For instructions, see Create an Alibaba Cloud account.

  2. After you create an Alibaba Cloud account, activate OSS.

  3. Optional: After you activate OSS, the default billing method is pay-as-you-go. To further reduce your OSS costs, we recommend that you purchase OSS resource plans. For instructions, see Resource plan purchase guide.

  4. You can use OSS through the console, GUI tools, command-line tools, and SDKs for various programming languages. For instructions, see Quick start.

• What compliance certifications does OSS have?

  Alibaba Cloud has achieved a comprehensive set of compliance certifications. It is a pioneer in obtaining major international compliance certifications in Asia, Europe, and other regions, as well as passing key assessments and security reviews in China. Alibaba Cloud is the first in the world to receive ISO 22301, CSA STAR Gold, and Germany's C5 additional clause certifications, and the first in Asia-Pacific for C5 and ISO 27001. It is also the first in China for MTCS Level 3 and ISO 20000. In 2012, Alibaba Cloud became the first cloud computing service provider in China to be audited and approved for ISO 27001 by BSI. Alibaba Cloud Object Storage Service (OSS) meets the compliance requirements of the U.S. Securities and Exchange Commission (SEC) and the Financial Industry Regulatory Authority (FINRA). Alibaba Cloud is also the first cloud vendor in China, following AWS, Azure, GCP, and IBM, to pass the audit certification by Cohasset Associates. For details, see Compliance certifications.

• Why is data restoration necessary?

  Cold data is data that is rarely accessed but must be stored for a long term. OSS uses underlying technologies to store this cold data at a very low cost. When needed, this data can be restored to an accessible state through a process called data restoration. This feature optimizes costs by trading the convenience of real-time access for significantly reduced storage expenses.

Alibaba Cloud regions

• Where is my data stored?

  When you create an OSS bucket, you specify an Alibaba Cloud region. By default, OSS stores your data in one availability zone within that region. If you enable zone-redundant storage, OSS stores your data across multiple availability zones within the same region for redundancy. This ensures that your data remains accessible even if one availability zone becomes unavailable.

• What is an Alibaba Cloud region?

  An Alibaba Cloud region is a geographical area that contains multiple physically separate and isolated availability zones. These availability zones are connected by a low-latency, high-throughput, and highly redundant network.

• What is an availability zone?

  An availability zone is a physical location within a region that has independent power and networking. Instances in the same availability zone experience lower network latency. Availability zones within the same region are interconnected through internal networks and are designed to be isolated from failures in other availability zones.

• How do I decide which Alibaba Cloud region to store my data in?

  When you select a region, we recommend that you consider factors such as geographical location, relationships between cloud products, and resource pricing. For more information, see How to choose an OSS region.

Billing

• How much does OSS cost?

  With OSS, you pay only for what you use with no minimum fees. You can also purchase resource plans in advance, and subsequent usage will be deducted from the corresponding credits. In most cases, resource plans are more cost-effective. For detailed pricing information, see the OSS pricing page.

• How am I billed if another account accesses my OSS resources?

  When another account accesses your OSS resources, you are charged according to standard OSS pricing. Alternatively, you can configure your bucket to use the pay-by-requester mode. In this mode, the requester pays for the requests and data downloads from your bucket. For instructions, see Pay-by-requester.

• How do I deactivate OSS?

  Because a one-click deactivation could impact your business, OSS does not currently offer this feature. You can use alternative methods to deactivate the OSS service or stop billing. For more information, see How do I deactivate OSS or stop OSS from charging my resources?

Data security and protection

• Is my data secure in OSS?

  OSS is secure by design. By default, only the owner of a newly created bucket can access its resources. OSS uses user authentication to control data access. You can use various access control policies, such as bucket-level and object-level access control lists (ACLs), to selectively grant permissions to users and user groups. The OSS console shows which of your buckets are publicly accessible. You can set buckets and objects that you do not want to be publicly accessible to private. If you set the ACL of a private bucket or object to public-read or public-read-write, OSS issues a warning. For more information about OSS security, see the OSS Security and Compliance White Paper.

• How do I control access to data stored in OSS?

  OSS provides multiple methods for controlling access to objects in a bucket, including ACLs, RAM policies, and bucket policies. For more information, see Overview of permissions and access control.

• What data encryption methods does OSS provide?

  Server-side encryption: When you upload an object, OSS encrypts it upon receipt and stores the encrypted data. When you download the object, OSS automatically decrypts it before sending it to you. The HTTP response header indicates that the object was encrypted on the server side. For more information, see Server-side encryption.

  Client-side encryption: You encrypt your data locally before uploading it to OSS. For more information, see Client-side encryption.

• How can I prevent data in a bucket from being accidentally deleted or overwritten?

  Versioning is a bucket-level data protection feature. When you enable versioning, overwrite and delete operations create historical versions of an object instead of permanently removing data. If you accidentally overwrite or delete an object, you can restore it to any of its previous versions. For more information, see Versioning.

• What is a retention policy?

  OSS supports the Write-Once-Read-Many (WORM) feature, which allows you to store data so that it cannot be deleted or modified. You can set a time-based retention policy for a bucket. Once the policy is locked, you can upload and read objects in the bucket, but no user can delete the objects or the policy until the retention period expires. Objects can be deleted only after their retention period ends.

  When you need to store important data for the long term and prevent it from being modified or deleted, such as medical records, technical documents, or contracts, you can store this data in a specified bucket and enable a retention policy to protect it.

• Does OSS support modifying objects online?

  OSS does not support online modification of uploaded objects. If you need to modify an object, you must first download it to your local machine, modify it, and then re-upload it.

• Does OSS use three-way replication?

  No. OSS uses erasure coding (EC) instead of three-way replication. Erasure coding provides performance and reliability comparable to three-way replication.

• How is the 99.995% availability calculated?

  The definition of the service availability SLA for OSS differs from that of instance-based products. Service availability is calculated based on the average of the error rates for every 5-minute interval within a service cycle. The calculation is as follows:

  • 5-minute error rate = (Number of failed requests in 5 minutes / Total number of valid requests in 5 minutes) × 100%

  • Service availability = (1 - Sum of 5-minute error rates in a service cycle / Total number of 5-minute intervals in the service cycle) × 100%

  For more information, see Object Storage Service (OSS) Service Level Agreement.

Data replication

• How do I replicate data from a bucket to another bucket in a different region?

  When you want to maintain an exact copy of the data in a bucket in a different region, you can configure one or more cross-region replication rules for that bucket. Cross-region replication automatically and asynchronously replicates objects across buckets in different OSS regions. It replicates object creation, update, and deletion operations from a source bucket to a destination bucket in another region.

• What are the use cases for cross-region replication?

  • Compliance: Although OSS stores multiple replicas of each object on physical disks by default, some regulations require data copies at a significant geographic distance. CRR replicates data between distant OSS regions to meet these requirements.

  • Reduced latency: Maintain object replicas in OSS regions closer to your users to minimize access latency across geographic locations.

  • Backup and disaster recovery: Store a copy of all data in another region to prepare for major disasters. If one data center is destroyed, activate the backup in the other region.

  • Data migration: Replicate data between OSS regions for business continuity or migration.

  • Operational needs: Keep object replicas in multiple regions for compute clusters that analyze the same dataset.

• How is cross-region replication billed?

  After you enable cross-region replication, the following fees may apply:

  • Storage fees: After data is replicated from the source bucket to the destination bucket, storage fees are incurred for the data in the destination bucket.

  • Cross-region replication traffic fees: When data is replicated from the source bucket to the destination bucket, data transfer traffic fees are generated. Cross-region replication traffic is billed on a pay-as-you-go basis only; resource plans are not available for this traffic.

  • Request fees: For each object synchronized, OSS accumulates the number of requests and bills them on a pay-as-you-go basis.

  If Replication Time Control (RTC) or Transfer Acceleration is enabled, corresponding fees also apply.

Data query

How do I query data?

OSS SelectObject allows you to use simple SQL expressions to retrieve specific data from objects in CSV or JSON format without retrieving the entire object. This feature simplifies the process of scanning object content and filtering it into a smaller, more targeted dataset. It is suitable for scenarios such as querying parts of large files, querying JSON files, and analyzing log files. For more information, see Query objects.

Storage management

• What is OSS lifecycle management, and how can it help reduce my storage costs?

  Lifecycle rules allow you to periodically transition less-frequently accessed data to the Infrequent Access, Archive, Cold Archive, or Deep Cold Archive storage classes, and delete data that is no longer needed. This helps you manage your stored data more efficiently, saving significant manual effort and storage costs. For example:

  • A medical institution's records are occasionally accessed within six months of being uploaded to OSS but are rarely accessed thereafter. You can set a lifecycle rule to transition medical records to the Archive storage class 180 days after they are uploaded.

  • A company's customer service call recordings are needed for statistical analysis and verification within the first two months after being uploaded to OSS. They are occasionally accessed after two months, rarely after six months, and no longer need to be stored after two years. You can set a lifecycle rule to transition the recordings to Infrequent Access after 60 days, to Archive after 180 days, and then delete them after 730 days.

  • A bucket contains a large number of objects that need to be deleted, but manually deleting more than 1,000 objects at a time can be cumbersome. In this case, you can configure a lifecycle rule that matches the entire bucket and sets all objects to be deleted after one day. All data in the bucket will be deleted the next day.

  For more information, see Lifecycle rules based on the last modified time.

• How do I periodically get information about the objects in my bucket?

  The OSS bucket inventory feature generates daily or weekly reports on specified objects in a bucket. This information includes the number of objects, their sizes, storage classes, and encryption status. For more information, see Bucket inventory.