Managed
One of the biggest obstacles sites and businesses are faced with is having the necessary solutions in place to accommodate these requirements along with the room for growth. This is easily handled by the right server solution, whether it be load balancing a cluster of web servers or a redundant fail-over setup for a mission-critical site. However, solutions like this can be quite expensive to setup in-house. These costs can be greatly reduced by allowing our skilled experts at QuexSolutions to manage your server setups.
Our team will work with you one-on-one to come up with a custom solution that meets your specific needs and budget. This is only after we have strategically planned the appropriate CPU, memory, disk I/O, concurrent users, database intensity, traffic surges, and of course future growth. The level 3 technicians work only with advanced setups. This means the same technicians will manage and monitor your setup, this is a very crucial aspect as it’s important that an expert who is familiar with the setup should also be the one working on it.
Feel free to contact us to get a quick quote on what would work best for you, and what it would cost as well.
Load Balance Your Sites Across Multiple Servers
Load balancing is one of the most commonly used solutions for high-traffic sites or resource-intense applications. The factor that makes load balancing so effective is that it disperses all incoming traffic to a website among a pool of servers, using a dedicated load balancer in front of the servers. In the event that a web server goes offline, it is immediately taken out of the cluster so none of your visitors are affected by the downtime.
There are many different ways to have a load balancer setup. So, therefore it’s important to evaluate each website or application’s requirements individually. Some setups may require high availability with redundant A+B network feeds and redundant load balancers. Others may need databases, webservers, and storage servers all separated. So it’s important to analyze your sites or applications needs and build accordingly.
Setup Details:
- Hardware Load Balancer – Kemp LoadMaster – 3000.
- Distributes incoming traffic for peak performance.
- Automatically bypasses any failed server, uses other servers to keep service online.
- Add additional servers easily without downtime.
- The same account manager and the dedicated team manage and monitor your infrastructure daily.
Web Server + Database Server
One way to have your site setup is to have the web server and database server separated into their own dedicated servers. This is great for a number of reasons; it separates the two primary services allowing each to run without resource contention. Another benefit is security, the database can be separated from the web server and made only privately accessible from the web server by secure local communication.
This setup is usually recommended if you have a database-intensive site. The web server and database server requirements are often different so this setup would allow each server to be setup and configured specifically for the needs and benefits of each. A single server with a very resource-heavy web server or database can affect each other’s performance, so having them separated can make a site that was once sluggish perform quickly again.
Features
- Separate Web Server and Database Server for optimal performance.
- Separate servers prevent contention for the same server resources (CPU, Memory, I/O, etc.).
- The database server is not publicly accessible, keeping the database extremely secure.
- Ability to vertically scale each platform based on how each grows.
- Web Server communicates to Database Server by local private connection.
Datacenter goes down? Your site won’t!
Geographic server failover is basically just DNS failover with monitoring. Essentially there is a monitor that checks the connectivity of your servers. If it detects an issue with it, the DNS for that server will be dropped and the short TTL (time to live) for the DNS will allow all traffic to propagate to the server that is working. Depending on what the TTL is set to, some users may only see a few minutes of downtime rather than the hours it would take to diagnose and fix a failed server.
Using a failover setup like this gives more peace of mind when it comes to reliability. It allows an entire ISP, data center, network, or server to fail, all while your site will automatically bring itself back online on a secondary server.
Features
- Two servers are located in separate data centers for the highest redundancy.
- Server failover only takes minutes, depending on what the Time To Live is set to.
- The site stays online while the failed server or network is fixed.
- Ideal for sites requiring high uptime.
- For high-end sites, server cluster failover is possible.
Storage Attached Network (SAN)
A SAN setup is ideal for multiple purposes. It is commonly used for a cloud cluster setup, where all servers use the SAN for storage and if a server fails or falls out of the cluster then all the data is still stored in a single location. Another useful purpose is when storage performance is crucial and there are multiple web servers. The SAN acts as a unifying storage point for the web servers and could consist of high-speed Solid State Drives (SSD) in raid for incredible performance and speed.
Using a SAN is beneficial as it cuts down on hard drive I/O and sync traffic between servers that are in a cluster. By having all data stored in a central location, bringing new servers online in various style setups is simple and easy. It’s a wise choice for horizontally expanding clusters.
Features
- Setup as a central storage location for multiple servers.
- High-performance SSD SAN’s for intensive workloads.
- Ability to add storage as needed for growth.
- Single RAID SAN’s or multiple for redundancy.
- Build and grow SAN(s) from 2TB to 120TB.
