Stackable Switches and Synology High Availability for Small Business Continuity Planning

avatar

Overview

This article, intended for IT Professionals, will discuss another enterprise level networking technology made accessible for small business, which is the ability of Stackable Switches. The ability of stacking switches can be found with the new “core aggregation” class of switch, where it features many 1GbE network ports, with a couple of 10GbE network ports which can be used for core aggregation of services, or can be used with another switch for increasing network redundancy.

Coupling Stackable Switches with Quad-NIC Synology products, such as the recently released DS2015xs, can effectively create a redundant network layer and redundant storage layer. The Network Layer is made redundant with Stackable Switches, and Synology High Availability will make the storage layer redundant.

This level of redundancy was typically wielded by large businesses with large budgets, where these businesses cannot afford downtime. But with today’s affordability of computer technology, these features are now being made accessible to small business. Today small business can enjoy the same functionality of large business, to minimize downtime should a critical failure occur at the network layer or storage layer.

What is a Stackable Switch?

A stackable switch is a type of network switch that is capable of operating together with one or more network switches. This group of switches will logically appear on the network as a single switch but having the port capacity of the sum of the combined switches.

For example, assuming 2x [48-service ports and 2-stacking ports] switches are configured for stacking, depending on the network deployment to be decided by the network administrator, this stacked configuration can either serve 96 single-network port devices for maximum served aggregation, or 48 dual-network port devices for maximum served aggregation with at least 1-port network redundancy.

With the configuration of the dual-network port per device, this allows the device served to incur a network switch failure and still maintain availability to the network. At the network layer, this is achieved by stacking the two switches together, and enabling Link Aggregation on both Switches, for the specific ports in use. For the “Application Server” using this redundant connection, it must be able to support bonding its physical network ports for failover, such as in VMware vSphere or Microsoft Windows Server 2012.


(Click to Enlarge)

Referring to network diagram above, a minimum of a Synology Quad-NIC capable NAS is needed, along with the stackable switches. In this example, the DS2015xs is being utilized as it supports on-board 2x 10GbE, which is being utilized by Synology High Availability to handle the heartbeat and data sync between the two nodes.

Looking at the DS2015xs itself, it can be seen where the 2x10GbE “Heartbeat/ data-syncing connection” has been bonded together using 802.3ad Link Aggregation (LAG). This allows for the heartbeat connection to be redundant against a cable failure. Looking at the remaining ports, (port 3 and 4 in this example), these have been bonded together with LAG, and that the physical network path is redundant with the two switches.

This configuration yields the network layer is made redundant, the storage layer is redundant, and for this scenario, the client computer has a redundant path to the network.

Application Scenario: Front-End Redundancy


(Click to Enlarge)

In this example, stackable switches along with Synology High availability are providing an example of “front-end redundancy”, where file-level applications, including file sharing over AFP, FTP, HTTP, NFS, SMB/CIFS, or WebDAV, and be made redundant against critical hardware failure. This configuration will yield that file-level services is still accessible for the business even if one of the switches has failed.

Note that the front-end Work Stations only feature a single NIC. Should a switch fail, the computers that were being served by that particular switch will no longer have access to the network. This can be remedied by deploying redundant network connections for each computer.

Application Scenario: Back-End Redundancy


(Click to Enlarge)

In this example, a form of “back-end redundancy” is being displayed, where block-level applications, including iSCSI or NFS for VMware vSphere, Microsoft Hyper-V, Citrix XenServer can be made redundant against critical hardware failure. This scenario is also taking advantage of the ability of clustering Application Servers together, to provide redundancy against application server failure. It also is taking advantage of dual network paths which are redundant.

Supporting dual redundant network paths supports the ability of MPIO for iSCSI Services. Supporting MPIO allows for increased network redundancy and load balancing on the back-end as the Application Server has two paths to use to pass iSCSI Data back to the storage array.

Conclusion

Network redundancy (whether on the front-end or the back-end) aims to achieve is to minimize a single point of failure that can affect an entire business. An end-user workstation with a network cable failure, yes, one person is affected. But if a network switch failed or the storage array experienced multiple drive failure, then the entire business is affected.

As technology becomes more affordable and accessible, a Small Business can begin to leverage technologies that were once limited to large businesses with large budgets. In previous years without this enterprise technology, should a small business experience a critical failure, often that business would experience downtime of a few days to attempt to recover from that failure. Downtime of a few days to rebuild/replace part of a network can cost a business thousands of dollars in lost revenue, and not to mention, a tarnished reputation.

For a small business which is leveraging stackable switches with Synology High Availability, this network eco-system is now redundant with a redundant network layer and a redundant storage layer. Businesses that take advantage of this technology can minimize their downtime should a critical failure occur within their network eco-system, maintaining business continuity and keeping their business in operation.

Further Reading

    1. Synology Blog: 10GbE Core Aggregation Switch Technology for Small Medium Business Storage, the XS+ with Intel SSDs
    2. Synology DiskStation DS2015xs
    3. Synology KB: How to set up Link Aggregation on Synology NAS
    4. Synology Blog: High Availability – System Failover for Small Businesses
    5. Synology KB: How to Use iSCSI Targets on Windows Computers with Multipath I/O