How Cloud Storage Affects File Management

Introduction
1.1. What is Cloud Storage?
Cloud storage is a computer data storage strategy in which digital data is kept in logical pools that are referred to as “the cloud.” A hosting business often owns and manages the physical environment, which consists of several servers (occasionally in multiple locations). These cloud storage providers oversee maintaining the data safe, secure, and operational, as well as the physical environment. People and businesses buy or lease storage capacity from suppliers to store user, organization, or application data. [1]
There are three types of cloud data storage.
- Object Storage — The huge scalability and metadata features of object storage are frequently utilized in cloud-based software. Data from other sources can be imported for backup, analytics, or archive applications.
- File Storage — Some applications require a file system to access shared files. A Network Attached Storage (NAS) server is frequently used to support this form of storage.
- Block Storage — Databases and ERP systems frequently need dedicated, low-latency storage for each host. Direct-attached storage (DAS) or a Storage Area Network (SAN) are examples of this.
Nowadays, there are plenty of cloud storage service providers, such as Google Cloud Storage, Amazon S3, DropBox, iCloud, Microsoft One Drive, Mega, Box, etc. There are four types of deployment options each provides us with different types of solutions to different types of needs.

- Public Cloud Storage — Customers who need to use computer resources, including hardware and software, can use public cloud storage. Public clouds are typically used for non-critical operations like file-sharing or application development and testing. Therefore, small-scale businesses without sensitive data use them. The public cloud can also be accessed by given to powerful anonymous users.
- Private Cloud Storage — Private cloud storage systems can be managed by the company directly or by a service provider outside the company. For example, private clouds are used by government agencies, financial institutions, and healthcare organizations to store sensitive and private data. It is possible for these private cloud infrastructures to be hosted on-site or by a different service provider.
- Hybrid Cloud Storage — A hybrid cloud storage system combines the concepts of private and public clouds. When adopting hybrid cloud storage, the company’s own cloud is used to store critical data while the public cloud is used to store the remainder. This approach is frequently used when an organization’s IT infrastructure needs to be quickly upgraded. A hybrid cloud is helpful to many businesses due to its versatility.
- Community Cloud Storage — This distribution approach is not accessible to everyone in need because it is targeted at many organizations within the same community. Even though other businesses use it, it is not private. Community clouds can be found at places like banks, public universities with shared spaces, and local police departments.
1.2. How Cloud Storage Affects File Management
In an organization, employees need to store and access documents every day. They should have a proper document management system to ease their work. Sometimes the allocated storage for a user in a document management system may not be enough for him. There should be additional storage for them to access in such a situation. For that, we can use cloud storage as a solution.
Advantages of using Cloud Storage
- Documents can keep proper version controlling system
- Cloud documents are easier to share between teams and colleagues, and can also be worked on simultaneously
- The cost of having additional storage is reduced
- Having good security for documents
- Files can be recovered when a disaster happens
Disadvantages of using Cloud Storage
- Difficult to decide what to migrate and what not to
- It may not be possible to migrate all files to the cloud at once
- Some sensitive documents can be easily raised
- Moving documents to the cloud can be a significant expense
Bucket Storage
2.1. What is Bucket Storage?
The Buckets resource represents a bucket in Cloud Storage. There is a single global namespace that includes all buckets. The most fundamental data storage containers are buckets. We need a bucket for each item we maintain in cloud storage. Although buckets cannot be nested like directories and folders, they can be used to organize data and control access to it. When we construct a bucket, we must give it a name that is internationally distinctive and the location where the bucket and its contents are kept. After it has been formed, the bucket’s name and location cannot be changed, but we may remove and recreate it to achieve the same result.
Objects are stored in the buckets. Individual pieces of data that we save in Cloud Storage are referred to as objects. In a bucket, there is no limit to the number of objects we can create.
Objects are made up of two parts:
1. Object data — a file that we want to save in Cloud Storage and is fully invisible to Cloud Storage.
2. Object metadata — a set of name-value pairs that define different aspects of an object.
Here are some object storage service providers.
1. Amazon Simple Storage Service (S3)
2. Google Cloud Storage
3. Azure Blob Storage
4. Backblaze
5. DigitalOcen Spaces
6. Qumulo
7. Object storage
8. Wasabi Cloud Storage
9. Amazon S3 Glacie
10. Oracle Cloud Infrastructure Object Storage Classic
According to the above information, Amazon Simple Storage Service (S3) and Google Cloud Storage are selected for further comparison. [2]
Amazon Simple Storage Service (S3)
- Can create only 100 buckets per account. According to our needs, we can increase up to 1000 buckets by submitting a service limit increase.
- The maximum bucket size is 5TB
- The maximum number of parts per upload is 10,000.
- An individual part has a maximum size limit of 5 GiB
- An individual part has a minimum size limit of 5 MiB unless it’s the last part, which has no minimum size limit
- Multipart upload API tracks the pieces
- Object-level tagging supported since Nov 2016
- DELETE without version specified applies DELETE marker to master. You still get the latest version of an object if the master is deleted.
- Flexible and dynamic control with Cross Region Replication API
- The cost is low. $0.06/GB at 90TB.
Google Cloud Storage
- There are no limits for creating buckets.
- The maximum bucket size is 5TB
- A multipart upload can have up to 10,000 parts.
- An individual part has a maximum size limit of 5 GiB
- An individual part has a minimum size limit of 5 MiB unless it’s the last part, which has no minimum size limit
- Multipart upload application needs the logic
- Object-level tagging is not available
- DELETE request without version moves from ‘master’ to ‘archive’. There is no concept of ‘version stack’.
- Data is stored redundantly with Multi-Regional Storage in a fixed manner.
- The cost is high. $0.08/GB at 90TB.
Conclusion
When considering the comparison between Amazon Simple Storage Service (S3) and Google Cloud Storage, we can get an idea that the two of them have some similar features. Even though we can see there are more advantages to Amazon Simple Storage Service (S3) has more than Google Cloud Storage, it can create only a limited number of buckets. Since we are not able to guess how many people will use document management systems throughout all the years that they will be used, we should consider Google Cloud Service. Because there is no limit for creating buckets in Google Cloud Storage.
References
[1] https://en.wikipedia.org/wiki/Cloud_storage
[2] Best Object Storage Software (2021–08–04)