Sentris.Com - Our Network

It also offers state of the art redundant electrical power and HVAC systems as well as multi layered security and CCTV systems, along with uninterruptible and back-up power, raised flooring, redundant Internet connections with 24-hour monitoring, and state-of-the-art fire suppression and environmental controls for your ultimate peace of mind.

Our 387,823 square feet data center can host over 200,000 servers and is monitored by our 24x7 on-site tech staff.

Internet Connections

On the Internet, speed is king. Slow download pages turn your users off. You get better return from your advertising investment when your pages download fast.

According to Boardwatch magizine, most of the Internet's traffic is carried by a few major backbone providers such as UUNET (MCI Worldcom), Sprint, AT&T, ...etc. To ensure our hosted web pages get to users fast, our data center is directly linked to these Internet backbones. Most users can reach Sentris data center via a single backbone which avoids the common network problem: traffic jam at the backbone intersections.

We personally running BGP with a blend of PCCW, NTT Verio and Level(3) with single-mode fiber connections, as well as interconnection to other Networks via the Seattle Internet Exchange (SIX). Soon, we'll add Savvis and Cogent paid peering. You can choose which of the 3 backbones you prefer as your primary provider, all are backed by each other in case of one of them goes down. And each backbone connect to their own cisco router to give you best redundancy. And each of the switches we use has a GigE uplink so you can get the most burstable speed.

For outbound, we look at each of the 250,000 destinations each of our transit providers send us and pick a path based on several criteria and pick the best route. If we get a route via peering, we always take it as those peers generally provide a shorter path than sending it to transit.

Our state-of-the-art Internet data center is engineered for high performance and redundancy. It has rich fiber connections to all major carriers and scalable bandwidth capacity from OC3 to OC192.

Network Operations

Our Network Operations Center is staffed 24x7 by Unix Systems Administrators, Microsoft Certified Systems Engineers, hardware experts, network engineers, programmers, and applications specialists. Our engineers monitor more than 900 servers, routers, switches and fiber-optic connections within the primary of our Network Operations Center located just inside our SDC in Seattle, Washington.

We have spent 2 years developing custom made monitoring and control software which will alert our NOC engineers by cellular phone, 2-way pager and email whenever a problem is detected by the auto-scripts. Often, a problem with the Server is corrected within minutes, if not seconds, from the time it occurred. Our proprietary software monitors the following ports and automatically restarts each service if it is not detected within 10 seconds: web traffic (port 80), POP3 email (port 110), outgoing email (port 25), telnet (port 23), ftp (port 21), secure server (port 443), real audio (port 8080 and port 7777), and other vital connectivity ports. Every 10 seconds our engineers get print-outs of any ports that did not respond and that could not be restarted automatically.

We have taken every step possible to ensure continued and uninterrupted operation of our servers: multi-homed single-mode fiber Gbits connections to diverse backbone providers, diesel-powered generators, redundant UPS systems, and on-hand inventory of hard drives, memory chips, network cards, motherboards, power supplies, processors, RAM and other hardware components.

You can rest assured that your server is being watched by the best of the best 24 hours a day.