Differentiating Performance Tests Versus Real Network Performance

Three areas require classification:

• What comprises a performance test?

• How does one interpret results?

• How does one compare the results to any realistic performance requirements?

The more common performance tests include blasting traffic from an input port to an output port of a device. For a given device, injecting traffic through multiple input ports to multiple output ports on the same device gives aggregate performance numbers. Usually, these tests are performed on Ethernet because Ethernet-based testers were the first available. Aggregate performance numbers are media-independent, but the type of media used plays an important role in defining what the theoretical packet-per-second limitation is. Table 5-4 shows characteristics of some of the more common media in use today.

Table 5-4 Media Characteristics

Interfram e Gap

Minimum Valid Frame

Maximum Valid Frame

Bandwid th

Ethernet

96 bits

64 bytes

1518 bytes

10 Mbps

Token Ring

4 bits

32 bytes

16K bytes

16 Mbps

Fiber Distributed Data Interface (FDDI)

0

34 bytes

4500 bytes

100 Mbps

Asynchronous Transfer Mode (ATM)

0

30 bytes (AAL5)

16K bytes (AAL5)

155 Mbps

Basic Rate Interface (BRI)

0

24 bytes (PPP)

1500 bytes (PPP)

128 Kbps

Primary Rate Interface (PRI)

0

24 bytes (PPP)

1500 bytes (PPP)

1.472 Mbps

T1

0

14 bytes (HDLC)

None (Theoretical) 4500 (Real)

1.5 Mbps

Fast Ethernet

96 bits

64 bytes

1518 bytes

100 Mbps

Calculating the theoretical maximum packets per second involves all the variables listed in Table 5-4: interframe gap, bandwidth, and frame size. The formula to compute this number is:

Bandwidth/Packet Size = Theoretical Maximum Packets per Second (where packet size may incorporate interframe gap in bits)

Table 5-5 lists the theoretical packet-per-second limitations for three common media—10 Mbps Ethernet, 16 Mbps Token Ring, and FDDI—each for eight different Ethernet frame sizes. These eight frame sizes, widely used in the industry, are derived from the performance testing methodology as outlined in the Internet standard for device benchmarking in RFC 1944. The numbers are derived by using the above formula.

NOTE RFC 1944 has recently been made obsolete by RFC 2544.

Table 5-5 Packet-per-Second Limitation

Ethernet Size (bytes)

10-Mbps Ethernet (PPS)

16-Mbps Token Ring (pps)

FDDI (pps)

64

14,880

24,691

152,439

128

8,445

13,793

85,616

256

4,528

7,326

45,620

512

2,349

3,780

23,585

768

1,586

2,547

15,903

1024

1,197

1,921

11,996

1280

961

1,542

9,630

1518

812

1,302

8,138

More specific detail in how the numbers in Table 5-5 were derived for the three media (10 Mbps Ethernet, 16 Mbps Token Ring, and FDDI) follow.

10 Mbps Ethernet The frame size needs to incorporate the data and header bytes as well as the bits used for the preamble and interframe gap, as shown in Figure 5-5.

Figure 5-5 10 Mbps Ethernet Frames

Direction of Data Flow

Preamble

Ethernet Frame #1

64 bits (8 x N Bytes, Where

Nis 18 Bytes Header + User Data)

Interframe Gap

96 bits

Preamble

Ethernet Frame #2

64 bits

In Figure 5-5 the fields have the following lengths:

• Frame—(8xN) bits (where N is Ethernet packet size in bytes, this includes 18 bytes of header)

16 Mbps Token Ring Neither token nor idles between packets are accounted for because the theoretical minima are hard to pin down, but by using only the frame format itself the maximum theoretical packets per second can be estimated, as shown in Figure 5-6. Because we are basing our initial frame on an Ethernet frame, note that we need to subtract the Ethernet header bits for the correct calculation of the data portion. So, for a 64-byte Ethernet frame, we get 64 - 18 = 46 bytes of data for the Data portion of the Token Ring frame shown in Figure 5-6.

Figure 5-6 16-Mbps Token Ring Frames

Direction of Data Flow <-

D

S

C

S

S

n

A

A

t

P

P

Vendor

Type

DATA

In Figure 5-6 the fields have the following lengths:

• Data—8x(N-18) bits (where N is original Ethernet frame size)

FDDI Neither token nor idles between packets are accounted for because the theoretical minima are hard to pin down, but by using only the frame format itself the maximum theoretical packets per second can be estimated, as shown in Figure 5-7. Note that, because we are basing our initial frame on an Ethernet frame, we need to subtract the Ethernet header bits for the correct calculation of the data portion. So, for a 64 byte Ethernet frame, we get 64 - 18 = 46 bytes of data for the Data portion of the FDDI frame shown in Figure 5-7.

Figure 5-7 FDDI Frames

Direction of Data Flow

Preamble

D

S

C

S

S

n

A

A

t

P

P

Vendor

Type

DATA

Preamble

In Figure 5-7 the fields have the following lengths:

• Data—8x(N-18) bits (where N is original Ethernet frame size)

Frame and Packet Size The packet size is a major factor in determining the maximum packets per second, and, in the theoretical test world, one packet size at a time is tested. Eight standard packet sizes are tested: 64-, 128-, 256-, 512-, 768-, 1024-, 1280-, and 1518-byte packets. Figure 5-8 shows a graph of the theoretical maximum packets per second for 10 Mbps Ethernet.

It is important to note that as the frame size increases, the maximum theoretical packets per second decrease.

Figure 5-8 10 Mbps Ethernet Theoretical Performance 18,000

Figure 5-8 10 Mbps Ethernet Theoretical Performance 18,000

64 128 256 512 768 1024 1280 1518 Size of Frames (Bytes)

64 128 256 512 768 1024 1280 1518 Size of Frames (Bytes)

Having seen how maximum theoretical performance is determined, we now see how that data fits in with the performance requirements of real user networks. Each medium has a specific fixed-size bandwidth pipe associated with it, and each one may or may not define a minimum and maximum valid frame size. The minimum and maximum frame sizes are important because most good applications written for workstations or PCs make efficient use of bandwidth available and use maximum-sized frames. The smaller the frame size, the higher the percentage of overhead relative to user data; in other words, smaller frame sizes mean less effective bandwidth utilization as illustrated in Figure 5-9.

Figure 5-9 Bandwidth Efficiency for Small Versus Large Frames

Bandwidth + Packet Size = Theoretical Performance

Smaller Packets (Less Efficient, Not Real)

Smaller Packets (Less Efficient, Not Real)

Larger Packets (Better Utilization)

Larger Packets (Better Utilization)

An understanding of real traffic patterns is important when designing networks. At least some typical applications should be known so that the average packet sizes on the network can be determined. Sniffer traces to look at typical packet sizes for varying applications are helpful; some of the more common ones include:

• Hypertext Transfer Protocol (HTTP) (World Wide Web)—400 to 1518 bytes

For optimal network designs, an understanding of the kinds of applications that will be used is necessary to determine the typical packet sizes that will be traversing your network. The following example, taken from a real network, shows how to optimize your network design.

0 0

Post a comment