Network Topology

You have been asked by the CEO of the hospital to aid IT in establishing a new network specifically for your department. You will link the MRI, CT, and Diagnostic Departments only. What sort of topology would you use? How would you construct this network, and why are you choosing to utilize this configuration?

After researching the benefits of the star topology, I would consider this to be a solid choice when linking the MRI, CT and Diagnostic departments together. According to Keary (2020), “Star topologies are most commonly-used because you can manage the entire network from one location: the central switch…if a node that isn’t the central node goes down then the network will remain up.” The type of security knowing that if one of the nodes goes down in MRI or CT, for example, is important because the network needs as little down time as possible. Adding new computers is also easy to do with this type of typology because the network doesn’t need to be taken offline. From a maintenance perspective Keary (2020) emphasizes that, “…star topologies require fewer cables than the other typology types. This makes them simple to set up and manage over the long-term”. By design, it is easier for network administrators to pinpoint faults in the network, therefore reducing any downtime. Although there is a great benefit to this typology, according to Cables to Go (2020), “The major drawback to this topology is that if the central device fails, then the network will become unstable or cease to function.”

There is an enhanced version of the star topology called the “extended star topology” (see picture below). According to Cables to Go (2020), “The extended star topology adds sub-central devices that are connect to the central device…The extended star topology is most suitable for large networks that may span an entire building.” These sub devices would reduce the amounts of long cable runs needed from the central device. The only long run would be to each department (MRI/CT and other diagnostics) and then a hub/switch would be used and each computer in that specific department would be connected to it.

To construct this network, I would employ one central web server with a backup web server in the center and depending on the length I would run either fiber optic or Category 5 RJ-45 cable to connect to sub-central switches which would be in each department. Each computer/device would connect to each sub-central switch. In order to secure the patient data I would employ a data server connected to the web server. This data server will act as the main storage and management of PACS. Each computer on the network will send/receive data from the main data server. I would have another data server running in parallel to act as a clone and also be connected to the web server. This would be only utilized in case the main data server for PACS goes down. The main internet service coming from the central web server would control the flow of data using a firewall which will allow only intranet sites to be accessed and restrict the flow of internet sites outside the network. A Wi-fi router in each department to be able to use mobile devices as well as Chromebooks for the medical personnel to use in conjunction with their desktop computer would be in place. Patient charts can be accessed online only through a web-based electronic medical record software either via desktop or Chromebook. This allows the patient flow (front desk Check-in, charting up, patient movement updates, imaging staff charting and finalizing) to be managed around the department at any location, not just the desktop computers. However, PACS will be only accessible on the hardwired network desktop computers in terms of sending imaging data to PACS.

I chose this type of configuration because my current employer utilizes something similar to my understanding. Although, I have yet to really dive into the details so I may have missed a couple of things. This network construction seems to be the most efficient, fastest and easily manageable configuration when compared to the other typologies I read about.

References

C2G. (2020). NETWORK TOPOLOGIES. Retrieved from Cables to Go: https://www.cablestogo.com/learning/library/data-center/network-topologies

Keary, T. (2020, June 4). Network Topology: 6 Network Topologies Explained & Compared. Retrieved from Comparitech: https://www.comparitech.com/net-admin/network-topologies-advantages-disadvantages/

Static vs Dynamic IP Addresses: Where to use and Why?

Let’s say you are setting up a network within the imaging department. You will utilize a network and also a WAN. What type of IP addresses will you use: static or dynamic? Why would you utilize these and where? Is this an important point to consider when setting up a network? Why or why not?

There are different types of IP addresses involved in networking such as static and dynamic. According to FS (2018), “A static IP address is an address that is permanently assigned to your network devices by your ISP, and does not change even if your device reboots.” A static IP is utilized for VPN, remote access, or hosting a website or server in a business setting. In contrast, a dynamic IP according to Vaughan-Nichols (2020) are, “…are subject to change, sometimes at a moment's notice. Dynamic addresses are assigned, as needed, by Dynamic Host Configuration Protocol (DHCP) servers.” Dynamic addresses exist because IPv4 has limited amount of static IP addresses. In a business setting the host website and/or web server would most likely have a static IP address but the individual devices (clients) that connect to the network (web server) would have dynamic IP addresses.

In an imaging department I would use static IP addresses for the web server, file servers, print servers and PACS system. Static IP addresses are better suited for these because according to there is better DNS support, server hosting (a static IP address makes it easier for users to find via DNS), more reliable connections (Vaughan-Nichols, 2020). One thing to note is that static IP address are more easily hackable which may require more network monitoring and security measures. Static IP addresses are pricier as well. Balancing reliability (connections) and security (hackability) would be one of the focus points for the IT department in a medical setting.

I would use dynamic IP addresses for the individual devices like Chromebooks and desktop computers that connect to the network. One reason is because according to FS (2018), “DHCP is advantageous for network administrators because it removes the repetitive task of assigning multiple IP addresses to each device on the network. It might only take a minute but when you are configuring hundreds of network devices, it really gets annoying.” This way there isn’t any conflicting IP address issues where multiple devices have the same IP. A wireless access point would be one device to use DHCP to auto-assign IP addresses to guest users (patients/visitors). However, a dynamic IP address would be unfit for a host website or server because DNS doesn’t work well when the address is always changing. This is important to note because this can lead to increased downtime which can cause havoc in a medical setting where data is constantly flowing between acquisition terminals, desktop computers and PACS.

References

FS. (2018, August 2). DHCP vs Static IP. Retrieved from FS: https://community.fs.com/blog/dhcp-vs-static-ip-differences.html

Vaughan-Nichols, S. J. (2020, April 22). Static vs. Dynamic IP Addresses. Retrieved from Avast: https://www.avast.com/c-static-vs-dynamic-ip-addresses#:~:text=Typically%2C%20static%20IP%20addresses%20are,usually%20fine%20for%20most%20consumers.