Servers are the backbone of businesses of all sizes. Mission critical applications and data reside on server hardware, and a single failure can be disastrous for the unprepared. If a server goes down, your business can come to a screeching halt. This is where a redundant server can save you.
Not only did an IDC study indicate that network downtime costs 80% of SMBs at least $20,000 (Source: TechRadar) but a different study indicated $1.7 trillion in total losses from data loss and downtime over the course of a year (Source: EMC). While these numbers can be hard to quantify with certainty, the implication is clear: downtime and data loss can cripple a business and businesses that find themselves without a redundancy at the server level are rolling the dice when it comes to their business continuity and data integrity. This is especially true for SMBs that struggle to find a balance between budget constraints and IT best practices. While forgetting about the “+1” in “N+1” can seem like an attractive idea when comparing sticker prices, prudent IT decision makers understand that server redundancy provides significant ROI when downtime mitigation, productivity, customer satisfaction, and reputation are considered.
In this piece, we’ll discuss some common SMB applications where a redundant server is a must, how an investment in a redundant server can be the difference between continued success and disaster, and how to select server hardware that is built with redundancy and data integrity in mind.
Active Directory is a great way for businesses to simplify administration, authentication, and access on a network. In Windows environments, it is common for users to depend on access granted through Active Directory for everything signing into a system to accessing email to accessing shared folders and drives. Should a server acting as an Active Directory Domain Controller go down without a redundant server available, employees won’t be able to access the systems they need to be productive. Not only will this have a direct impact on the customer and vendor relationships the employees help maintain, should downtime occur for an extended period of time or become a regular occurrence, employees can become disengaged and morale can be impacted.
To quote Microsoft “every domain in your Active Directory environment should have at least two functioning domain controllers." This recommendation should be followed for almost every production environment. In the case of Active Directory, an investment in a redundant server is an investment in productivity and employee engagement.
High-Availability Clustering for Virtual Environments
As virtualization becomes a reality for more and more businesses, it can become easy to overlook the need for redundancy at the hardware level. However, as any sysadmin who has ever had to rebuild a cluster of virtual machines can tell you, fault-tolerance at the hypervisor level is a very important part of IT operations. If you are running all of your servers as virtual machines on one physical server, you have created a single point of failure that could bring all productivity to a standstill, cause data loss and corruption, and take a significant amount of time to recover from should a failure occur.
You can mitigate a lot of the risk by implementing a second server and configuring the virtual machines on your hypervisors to run in one of multiple “high-availability” cluster configurations. Popular hypervisor vendors like VMware (ESXi) and Microsoft (HyperV) support various fault-tolerant and high-availability configurations that allow for automatic migration of virtual machines from one hypervisor to another when specified events or failures occur. A prerequisite to leveraging these features is a server to run the second hypervisor on. Running a hypervisor on two physical servers and configuring failover to allow for high-availability is a good way for a business with a limited IT budget to strike a balance between keeping costs down and critical services up.
Point of Sale and Front Office
Having a system go down in the middle of a transaction can significantly damage a business’s reputation in the eyes of the affected customers. For small offices (such as doctor’s offices, insurance agencies, accounting firms, or other professional offices) or retail locations that depend on back-office database servers, transaction processing systems (TPS), or point of sale (POS) systems, a server outage in the middle of the day can mean nothing gets done. As data lives on the server, in the event of a failure, not only will sales become a challenge, but so will pulling up client records to answer simple inquiries. Client’s opinions of the business can change as a result, and not only will the immediate transaction be in jeopardy, the long-term relationship can be negatively affected.
The challenge becomes even greater when you consider that these smaller offices and shops generally do not have dedicated IT staff available to bring servers back online quickly. This is a great example of an ounce of prevention being a pound of cure and the benefits of redundant servers. By implementing a redundant server, businesses can help maximize the amount of time they spend serving their customers and minimize the amount of time they spend scrambling to troubleshoot a downed server.
We live in a data-driven society and a large amount of the data businesses depend on resides in databases. While database backups are a vital part of any disaster planning and recovery plans, backups alone aren’t enough to maintain data integrity in environments where data is updated almost constantly, as is the case with many business databases. Installing a redundant database server and configuring database replication can be the difference between a server outage causing lost revenue and corrupted data or simply being a blip that can be addressed with the redundant server takes over.
For a detailed example of configuring database replication with MySQL, check out the MySQL documentation. The short version is: a master database server is configured with at least one slave server, data between the two is synced, and should the master fail, the slave can take over. Once the master is restored, it can be redeployed as the slave.
Selecting a Server
When selecting redundant server hardware, it is important to look for hardware that is built with reliability, fault tolerance, and serviceability in mind. For example, when architecting a smaller scale redundancy solution for a dental franchise, Premio partner Echostreams decided on the DSS212 from the Durastream line of mission-critical servers. The DSS212 was selected because its fault tolerance, redundant features, and storage capacity made it an ideal fit for the dental franchise’s application and budget (for more, see this Premio post).
For database applications, solid state storage may be the right balance of speed, performance. Premio’s Flachestream Solid State Drive Servers not only offer the benefits of solid-state storage technology, they are also built with hot-swappable components and many offer features like redundant power supplies and network interfaces to maximize availability and up-time.
In conclusion, server downtime and data loss are two of the biggest threats to business continuity. In the hyper-competitive world of business, a decision to forgo redundancy at the server level can prove costly. By making a modest upfront investment in redundant servers and selecting the right server hardware for their use case, businesses can mitigate their exposure to downtime and data loss. Contact us today to get architect a server or storage solution designed to meet the unique demands of your business.