If you're accessing your data anytime, anywhere in the cloud, location shouldn't matter, right?
As it turns out, it does. There are several reasons why it matters where your cloud storage is located:
Legal & Regulatory Policy: How do companies ensure they are archiving and protecting business data to comply with electronic data laws? According to BCS for example, no matter what data storage and security strategy an organization uses, IT decision makers should consider these six key questions:
Performance: To reduce network latency, cloud storage and the applications that access it should be as close together as possible, even in the cloud, and they need to be close to the end-user. Thus New York-based users who use NY-based applications should have their storage in a cloud in the NY area as well.
Backup & Replication: Cloud-based backup and recovery makes sense as well. Having multiple instances of your data replicated by geography is a key function for distributed datacenter replication, and shows potential for rapid growth.
So, at Mezeo, we see three ways to think about cloud storage and geographic options and how to improve the distribution of data across geographically distributed data networks:
Geo-Location: Locating stored objects close to where they will be used for. Faster access via the closest cloud storage instance using data center peering (this also allows you to define where you store your data/objects).
Geo-Replication: Replication through policies, with uninterrupted access to content.
Single Namespace: Providing a single means of access to stored objects regardless of where the objects are located.
Geographic placement supports creation of an object in a specific cloud storage instance. At Mezeo, our replication policy allows for the specification of the locations of the replicants. For example, the policy indicates "create the object in New York, LA, and Houston." If an object is created in New York, it will be replicated to LA and Houston. If it created in Houston, it will be replicated in New York and LA.
Some storage vendors support replication as a component of their disaster recovery recommendations. If your selected storage vendor offers this option, then the storage solution could ensure there are at least two copies of every object in every instance of Mezeo's cloud storage. Recovery in the case of disaster with this approach would be handled by the storage vendor's solution.
By considering a combination of replication provided by storage vendors and replication provided by Mezeo, a service provider could offer a highly differentiated service. Your customers would be assured of recovery in the case of any possible failure, from a single disk failure to a catastrophic data center loss. Mezeo works with our service providers to determine the benefits of various replication options and the impact as you design your SLA level(s).
Policies are assigned in the onboarding/provisioning process and may be updated if requirements change. There are also special situations for policy updates, such as if a particular data center has a catastrophic outage, the policies associated with replication to the Mezeo instance in that data center can be modified.
As it turns out, it does. There are several reasons why it matters where your cloud storage is located:
Legal & Regulatory Policy: How do companies ensure they are archiving and protecting business data to comply with electronic data laws? According to BCS for example, no matter what data storage and security strategy an organization uses, IT decision makers should consider these six key questions:
- Will content be stored and remain unaltered over the required retention time frame?
- How will this technology stay updated to ensure long-term availability of records?
- Does this technology enable the organization to retrieve data quickly enough to respond to a legal request within the stipulated deadline?
- Can this technology grow with the business and meet regulatory requirements?
- Can this technology be used with other content generating applications?
- How will this data storage architecture address litigation and discovery challenges?
Performance: To reduce network latency, cloud storage and the applications that access it should be as close together as possible, even in the cloud, and they need to be close to the end-user. Thus New York-based users who use NY-based applications should have their storage in a cloud in the NY area as well.
Backup & Replication: Cloud-based backup and recovery makes sense as well. Having multiple instances of your data replicated by geography is a key function for distributed datacenter replication, and shows potential for rapid growth.
So, at Mezeo, we see three ways to think about cloud storage and geographic options and how to improve the distribution of data across geographically distributed data networks:
Geo-Location: Locating stored objects close to where they will be used for. Faster access via the closest cloud storage instance using data center peering (this also allows you to define where you store your data/objects).
Geo-Replication: Replication through policies, with uninterrupted access to content.
Single Namespace: Providing a single means of access to stored objects regardless of where the objects are located.
Geographic placement supports creation of an object in a specific cloud storage instance. At Mezeo, our replication policy allows for the specification of the locations of the replicants. For example, the policy indicates "create the object in New York, LA, and Houston." If an object is created in New York, it will be replicated to LA and Houston. If it created in Houston, it will be replicated in New York and LA.
Some storage vendors support replication as a component of their disaster recovery recommendations. If your selected storage vendor offers this option, then the storage solution could ensure there are at least two copies of every object in every instance of Mezeo's cloud storage. Recovery in the case of disaster with this approach would be handled by the storage vendor's solution.
By considering a combination of replication provided by storage vendors and replication provided by Mezeo, a service provider could offer a highly differentiated service. Your customers would be assured of recovery in the case of any possible failure, from a single disk failure to a catastrophic data center loss. Mezeo works with our service providers to determine the benefits of various replication options and the impact as you design your SLA level(s).
Policies are assigned in the onboarding/provisioning process and may be updated if requirements change. There are also special situations for policy updates, such as if a particular data center has a catastrophic outage, the policies associated with replication to the Mezeo instance in that data center can be modified.
Leave a comment