next up previous
Next: About this document Up: New Delay Analysis in Previous: Decomposed Method

Service Curve Method

 

In Section 1 we informally introduced the service curve model. In this section we apply this method to derive a closed form for a lower bound on the worst-case end-to-end delay tex2html_wrap_inline2235 for Connection 0 in Section 4 using the service curve method.. We first derive a closed form for an upper bound on the service curve for a FIFO server, and then apply this result to derive the upper bound on the worst-case end-to-end delay lower bound on the worst case delay.

Let tex2html_wrap_inline2318 be the input traffic function of Connection i at Server k and tex2html_wrap_inline2324 be the constraint function of input traffic tex2html_wrap_inline2318 , and let tex2html_wrap_inline2328 be the output traffic of Connection i at Server k and tex2html_wrap_inline2334 be the constraint function of output traffic tex2html_wrap_inline2282 .

  definition1102

The service curve can be used to effectively evaluate the end-to-end delay suffered by traffic, as described in the following theorem.

theorem1116

Consider Server k in Figure 3 serving n+1 connections. Let

equation1136

and

equation1142

  lemma1145

Proof:

For any time t, the amount of data departing the server during [0,t) is tex2html_wrap_inline2376 , and tex2html_wrap_inline2378 is the time of the tex2html_wrap_inline2376 -th bit arrival at the server. Since the server uses a FIFO scheduling discipline, we know that tex2html_wrap_inline2382 bits of data from Connection i have departed the server during time interval [0,t). So

equation1166

Therefore, by the definition, we know that if

equation1174

then tex2html_wrap_inline2025 is a service curve of Connection i offered by Server k. Q.E.D

Let

  equation1185

and

  equation1194

The definition for service curve is very loose. For a connection, there are infinitely many functions that satisfy Inequality (C.26). For example, a function with zero value is a service curve by the definition. In the following, we define an approximation for the service curve of a FIFO server.

  theorem1207

Proof:

For any tex2html_wrap_inline2404 , if tex2html_wrap_inline2406 for tex2html_wrap_inline2408 , let tex2html_wrap_inline2410 , we have that

eqnarray1220

On the other hand, let tex2html_wrap_inline2412 for all tex2html_wrap_inline2414 , it is easy to know that

equation1232

and

equation1243

Therefore, we have that

equation1246

and

eqnarray1255

This is a contradiction to Lemma C.6. Q.E.D

corollary1276

Proof:

According to Theorem C.4, we have that tex2html_wrap_inline2424 . For tex2html_wrap_inline2426 , the maximum available service for Connection i during time interval [0,t) is tex2html_wrap_inline2432 . Therefore tex2html_wrap_inline2434 for all tex2html_wrap_inline2436 . Q.E.D

In order to use the above results, we need to accurately estimate the internal network traffic. Following Lemma C.6 gives a tight estimation about the constraint function of output traffic of each connection.

  lemma1303

Proof:

Let tex2html_wrap_inline2450 be the time such that tex2html_wrap_inline2452 . Without loss of generality, we assume that tex2html_wrap_inline2454 for tex2html_wrap_inline2456 and tex2html_wrap_inline2414 , tex2html_wrap_inline2460 for tex2html_wrap_inline2456 . After time tex2html_wrap_inline2450 , we assume that tex2html_wrap_inline2466 for tex2html_wrap_inline2468 and tex2html_wrap_inline2414 , tex2html_wrap_inline2472 for tex2html_wrap_inline2468 . Since at time tex2html_wrap_inline2450 , the queue length of Server k is tex2html_wrap_inline2480 , according to FIFO scheduling discipline, during tex2html_wrap_inline2482 , the output traffic of Connection i is zero. But after time tex2html_wrap_inline2486 , the output traffic of Connection j ( tex2html_wrap_inline2414 ) is zero. Hence

equation1349

and

equation1356

Therefore we have that

equation1368

Q.E.D

Now we can approximate the service curves offered by servers to Connection 0 in the example presented in Section 4.

theorem1376

Proof:

According to Theorem C.4, for k=1, we have

 

  

Figure 9: Connections at the First and the Second Server.

 

eqnarray1410

For k=2, we know that the traffic constraint functions for Connection 3 and Connection 4 are the same as tex2html_wrap_inline2502 . Furthermore, by Lemma C.6, the constraint function of Connection 2 is lower bounded by tex2html_wrap_inline2504 , where tex2html_wrap_inline2506 is the maximum queue length of the middle output port of the first switch when it only serves Connection 0 and Connection 1 and can be easy evaluated as

eqnarray1418

Substitute tex2html_wrap_inline2508 into tex2html_wrap_inline2510 , we have that tex2html_wrap_inline2512 . Therefore by Theorem C.4, we obtain that

eqnarray1437

 
Figure 10: Connections at the k-th switch.  

For k>2, similarly, we can obtain that

eqnarray1450

where tex2html_wrap_inline2516 is the maximum queue length of the middle output port of the (k-1)-th server when it only serves Connection 0, Connection (2k-3), and Connection (2k-4) and can be easy evaluated as

eqnarray1457

Therefore, we have

eqnarray1470

Q.E.D

According to Theorem C.4, network service curve tex2html_wrap_inline2524 of Connection 0 can be estimated as following:

eqnarray1476

Therefore

eqnarray1498


next up previous
Next: About this document Up: New Delay Analysis in Previous: Decomposed Method

Riccardo Bettati
Wed Jul 14 18:25:49 CDT 1999