Below are some suggestions to help resolve poor performance results when using a Lumen network service (based on many possible causes). The steps below are loosely‑ordered according to degree of potential difficulty, from easier to harder. To get help, contact Lumen at any time either by creating a repair ticket in Control Center or by contacting Enterprise and Wholesale Repair.
You can measure performance using testing programs or by using a file copy test. Two types of testing programs are mentioned in the sections below, a “bandwidth” or “speed” testing program for public Internet service testing (such as the Lumen Internet speed test) and an “Iperf” testing program for use on any network service. Perform on‑line searches for more information on bandwidth testing, speed testing, and Iperf testing.
Keep in mind performance measurements rarely achieve a full 100% throughput result, as reported by a “bandwidth” or “Iperf” testing program. Learn more about acceptable results
Your Lumen service is configured to deliver the transmission capacity ordered. See that order details and other communication from Lumen confirm the service capacity that you expected to receive.
The wireless transmission between a test computer and network equipment is often the weakest link within the overall transmission path. Try to position a test computer in a location where the WiFi signal is strongest.
If you are testing from a home or small office, it may be easy to restart the network and WiFi equipment. Unplug the equipment for a minute.
The network equipment providing the WiFi signal often includes standard Ethernet plug connections. Connect a laptop test computer directly using an Ethernet cable. If test results are improved when using a cable, you should consider upgrading WiFi equipment.
When you run the test, see that other applications are not active. In a home or small office, the activity light on network equipment shouldn't appear as a constant flicker.
Performance limitations within the test computer can affect a measurement result when testing higher‑speed service connections. See that the test computer is not running other applications. If available, try a newer or better test computer.
If you're using an Internet “bandwidth” testing program, try using a bandwidth test program from another provider. Results will vary, but should be consistent over time.
You or others may have access to equipment or software systems that record accumulated transmission errors. Check to see if new errors are occurring presently. Error counters can be reset to zero in many systems.
Is performance poor for one specific application or task, or do multiple unrelated activities all seem slow? Gain an informed understanding of the scope of the problem. In some cases, the geographic transmission distance to reach the application server or web site can affect performance. See data transfer latency.
Most Lumen service connections include monitoring equipment placed in the service‑entrance area of a property. Lumen can initiate remote testing to measure performance and errors. These same tests were performed on the date of service activation. Perhaps something has changed. Create a repair ticket in Control Center.
Some lower‑cost Lumen service connections may not use testing equipment placed in the service‑entrance area of a property. However, testing through to other on‑site equipment can be performed.
Your Lumen service hand‑off connection may include a segment of wire or fiber located within your building. The connection between Lumen equipment in the property service‑entrance area and the equipment within you own suite or space may be provided by a building owner, often working with a wire and fiber contractor. In some cases, the in‑building extension is provided by Lumen.
See that the connection between your equipment and the provided hand‑off interface is good. Added repeaters or cross‑connections can be a source of trouble. Excessive added distance can cause errors.
Further testing coordinated with Lumen may help isolate errors occurring in the building extension segment. To request additional testing, create a repair ticket in Control Center.
The performance of computer applications is also affected by transmission distance delay, the measured time (or latency) between a computer and a server. A data transfer test conducted between local‑area destinations can produce results quite different from the same test over longer inter‑city distances, even when the bit‑rate or “speed” capacity is the same.
Some testing programs perform a data transfer test, a measurement that includes the use of standard protocols to control transmission pacing. The Iperf testing program is an example. Iperf data transfer test results over long distance can be inaccurate if default settings are used within the program. The Iperf testing program includes adjustments for the TCP receive window size to account for the transmission distance (latency) between the client and server test computers. See Iperf instructions and Internet search resources. (Windows users should consider using ntttcp (Windows NT Test TCP) or ctsTraffic instead of Iperf.)
Note: A ”bandwidth” Internet performance testing program doesn't need any latency adjustment. These programs don't perform a data transfer test with standard protocols to control transmission pacing.
Latency caused by long distance can also limit performance of a standard file‑copying task when longer distances and higher network speeds are present. To some degree, the file copying features used will automatically perform pacing adjustments to account for longer distances (TCP). See that computers and servers are upgraded to the latest operating system versions.
Note: Higher‑performance file copying programs may be available from vendors.
Network router equipment may need adjustment to perform best when a Lumen service connection is configured to operate at an incremental speed. A purchased Lumen service capacity may be a smaller increment of 100 Mbps, 1000 Mbps, or 10000 Mbps speeds—the standard speeds of Ethernet equipment interfaces. In these cases, Lumen or partner metro‑area network equipment will use software settings to enforce the lower rate purchased, a traffic “policing” policy applied to each customer.
A data transfer measurement test, using a test program or file copying, may yield slower results when an incremental speed is purchased. Without added buffering controls to align with the purchased incremental rate, a router or other device may constantly send transmissions at the full‑interface speed, not the slower incremental speed. Elevated packet discards can occur, as enforced by the traffic policing policy. TCP pacing controls will reduce performance for data transfer operations.
You can adjust a router to operate at the lower incremental bit rate with added buffers to absorb brief bursts of traffic. Vendor router products may characterize this added configuration setting as a “traffic shaping” feature. The configured Ethernet traffic shaping rate should match the lower incremental speed of the service purchased. Packet discards will be reduced. Learn more about traffic shaping
Note: SD‑WAN devices may include dynamic traffic shaping adjustments; a manual setting adjustment should not be needed.
Consult router vendor product documentation or Internet search resources for a description of Ethernet traffic shaping features. A router managed by Lumen is adjusted by Lumen.
If a specific user task within an application is a primary performance concern and is used as a primary basis of testing measurement, you should apply any available upgrades of the application. You can also search online or contact the application publisher for any technical advice regarding response time optimization.
A Lumen technician may have other things to try when the remedies described above are unsatisfactory. There may be Lumen network equipment adjustments to consider when accommodating unusual applications or older systems. There may be a need to change or tune access path elements, including the elements of a local partner network when a service location is not served by Lumen directly.
You may see improved performance test results after completing the suggested remedies above. But performance measurements rarely achieve a full 100‑percent throughput result, as reported by a bandwidth or Iperf testing program. Added overhead protocols may be present and unavoidable. You should try multiple tests. For lower-cost services, test results can be more variable. You should consider an averaged test result within a high‑nineties range of the purchased service capacity as acceptable.
Perhaps a specific application or user activity remains slow even when test‑program results are good and all remedies have been applied. The network service may not be the cause of slowness. The server itself may be a congestion point. Internet web sites are often slowed by added advertising content. Similarly, a private application may be slowed by operations that interact with other servers, servers that could be overloaded and some distance away, perhaps administered by another company.
The internal design of a database application can also affect the response times for users. An application developed for use over a short‑distance within a building may struggle when long‑distance users are added. It may be true that an increase of network capacity provides no benefit. Developers may need to upgrade software or reconfigure the coding of retrieval operations to reduce repetitive network traffic exchanges.
It is important that your new Lumen service provides performance results that meet your expectations. For most Lumen products, your new service has already been tested during the activation process. Lumen is confident that all of the performance aspects controlled by Lumen are configured correctly for the service capacity ordered. However, there may be performance‑limiting aspects outside of what Lumen can know or control. To get help, contact Lumen at any time either by creating a repair ticket in Control Center or by contacting Enterprise and Wholesale Repair.