検索
言語

非同期レプリケーション

Protect your data at any distance

本番サイトでのデータ保護対策がすべて失敗する日が来るかもしれません。ダウンタイム1分ごとに、10,000〜100,000ドルの収益が失われ、さらに悪いことに、業界の評判が損なわれる可能性もあります。非同期リモートレプリケーションでは、本番サイトが利用できなくなった場合でも、データは複製され、すぐにアクセスできるようになります。非同期レプリケーションまたは非同期ミラーリングは、すべてのアプリケーションとビジネスクリティカルなデータに対して、最高の完全性と最も柔軟なディザスタリカバリソリューションを提供します。

One of the first issues to resolve is the location of your remote site. This can be either one of your facilities far from your primary data center, a distant co-location facility shared with others, or a public cloud service. DataCore SANsymphony offers remote replication solutions to satisfy any of these choices. The graphic on the right illustrates a common configuration where the production site operates on a fully-redundant synchronously mirrored pair of SANsymphony nodes replicating asynchronously to a remote SANsymphony node over a WAN.

DataCore SANsymphonyを示す非同期データレプリケーションの図
Asynchronous remote replication with DataCore SANsymphony ensures mirroring of data between sites, ensuring disaster recovery

Replicating to Azure Cloud

If you would like to take advantage of the scalability, flexibility and cost-efficiencies of cloud services and/or if you do not have an appropriate physical location available, DataCore™ Cloud Replication on the Azure Marketplace offers a convenient alternative.

DataCore-authorized partners can help you remotely replicate to Azure while being able to recover from anywhere in the world and scale your offsite infrastructure as needed. Other options include replicating to Amazon Web Services.

Dynamic Mirror Path Balancing to Increase Usage Density

Dynamic mirror path balancing allows the expansion and the ability to maximize the utilization of interconnections between DataCore SANsymphony nodes. This is done by increasing the functionality of the asynchronous mirroring and utilizing dynamic active pathing to improve the throughput and resiliency.

Advanced Site Recovery

The data is safe at the remote site, now what? Many companies go this far, but fail to take the next critical step in their DR plan: Testing. With Asynchronous remote replication, non-disruptive tests can be conducted against the production data and applications as often as needed. Simply engage test mode for any volume at the remote site and the remote application servers are presented with the latest copy of the production volume. When the test is complete, simply disengage test mode and the state of the system reverts back, with no impact on production.

But what if a failover and then a subsequent failback become necessary? This is much more difficult. DataCore SANsymphony’s Advanced Site Recovery feature automatically reverses the asynchronous replication direction after the production site is ready to re-synchronized with the latest updates that have transpired at the DR site. Once both locations are in synch, the production site can once again take over primary IT responsibilities. And, unlike other solutions, a full resynchronization of all the data is not necessary to bring the original site back online. This saves time, money, and a lot of frustration, not to mention prevents unnecessary downtime during the transition.

Site to Site Data Replication diagram
The Advanced Site Recovery feature automatically reverses the asynchronous replication direction after a failover or failback operation to ensure that the alternate site receives all updates once services are restored.

Asynchronous Replication Feature Highlights

  • Perfect for disaster recovery, inter-site migrations, and remote site application testing
  • Only needs a basic IP connection to the secondary sites
  • Bi-directional transfers
  • Replicate to remote plants, co-location facilities and public cloud services
  • Available on Azure Marketplace as DataCore Cloud Replication
  • Asynchronous protocol to tolerate transmission delays and link outages
  • Central site to remote sites (one-to-many)
  • Remote sites to central site (many-to-one)
  • Many-to-many
  • Compressed, multi-stream transfers for fastest performance and optimum use of bandwidth
  • Allocate link bandwidth among replicated disks based on user-defined storage profiles
  • Expedite initialization of remote site using transportable media
  • Test disaster recovery readiness without impacting production
  • Direction of replication can be reversed using Advanced Site Recovery
  • Integration with VMware Site Recovery Manager (SRM)

同期レプリケーションと非同期レプリケーション:両者はどう違うのか?

非同期データレプリケーションは、伝送遅延のために同期レプリケーションができない離れたストレージデバイス間でデータを複製します。たとえば、プライマリデータセンターから遠隔地のディザスターリカバリー拠点にレプリケーションする場合などです。データを書き込むアプリケーションは、リモートサイトでのコピーの保存を待つことなく、ローカルにデータが保存された時点で確認応答を受け取ります。

In contrast, with synchronous replication (also known as synchronous mirroring) hosts receive acknowledgment of writes only after the remote site has saved a copy. This ensures that both sites have the same data. If one site fails, the other site can seamlessly take over without disruption; a pre-requisite for uninterrupted access across stretch or metro clusters configured for high availability / business continuity.

Both techniques are used in tandem, proving complementary for different use cases. DataCore SANsymphony supports both asynchronous replication and synchronous mirroring empowering IT teams to serve their needs both in terms of disaster recovery and business continuity.

ソフトウェアディファインドブロックストレージであるSANsymphonyを使ってみましょう