1 Modellazione e valutazione di un ambiente applicativo su una intranet

2 Caratterizzazione del carico Applicazioni considerate sulla intranet: 1) Corporate training: web based application 2) Access to local file system

3 Ipotesi sulle infrastrutture: Tutti i file server e il web server hanno una singola CPU ed un singolo disco FDDI e il router sono molto veloci rispetto alla Lan Ethernet, quindi sono modellati come semplici ritardi (delay), quindi centro di servizio senza coda di attesa. Tutte le altre componenti sono modellate come code con serventi con tempi di servizio indipendenti dal carico

4 Ipotesi sugli utenti: numero degli utenti finito un utente genera una nuova richiesta, dopo aver ricevuto la risposta dalla precedente, dopo un ritardo pari al “think time” 85% degli utenti lavora con il file system locale 15% con il Web Server

5 FDDI 100 Mbps Lan 5 R1 R3 R2 R4 Lan 1 10 Mbps Eth Lan 2 10 Mbps Eth Lan 3 10 Mbps Eth Lan 4 16 Mbps TR File server 4 File server 2 Web server 100 Windows NT clients File server Windows NT clients File server 1 50 Unix Workstation Rete intranet considerata 100 Windows NT clients

6 Modello rete di code Modello chiuso multiclasse (client group, application, server) client group: CLi: clients in Lan i (i: 1 to 4) application: –FS for local file server access, –TR for Training server: –FSi: i-th NFS server (i: 1 to 4) –WebS: Web Server

7 Tipi di classi e numero di utenti (CL1, FS, FS1)120 x 0.85 = 102 (CL2, FS, FS2)50 X 0.85 = 43 (CL3, FS, FS3)100 x 0.85 = 85 (CL4, FS, FS4)100 x 0.85 = 85 (CL1, TR, WebS)120 x 0.15 = 18 (CL2, TR, WebS)50 x 0,15 = 7 (CL3, TR, WebS)100 x 0,15 = 15 (CL4, TR, WebS)100 x 0,15 = 15

8 Tipi di serventi (risorse usate) Routers: basso ritardo dovuto a bassa latenza FDDI ring: basso ritardo dovuto ad alta banda CPU Disks serventi con tempi di servizio indipendenti dal carico LANs Service Demands: D i,r = V i,r x S i,r where V i,r = Visit Ratio S i,r = Service Time

9 FDDIR3R1 R4 R2 L2 D D D D C C C C (CL1, Tr, Web) (CL2, Tr, Web) (CL4, Tr, Web) (CL3, FS, Fs3) L1L3 (CL1,TR, Web) (CL1, FS, Fs1) (CL2,TR, Web)(CL2, FS, Fs2) (CL3, TR, Web) (CL1,TR, Web) (CL2,TR, Web) (CL3,TR, Web) (CL4,TR, Web) (CL1, FS, Fs1) QN model Web S

10 Web server workload characterization (for a training session) Avg request document size per HTTP request –20 rqs for txt documents (2.000 bytes per doc) –100 rqs for inline images ( bytes each) (20 text pages x 5 inline/text pages) –15 rqs for other multi-media (mm) obj ( bytes each)

11 Web server workload characterization % request for: –txt documents = 20/( ) = 15 % –inline images = 100/( ) = 74 % –other mm obj = 15/( ) = 11 %

12 Web server workload characterization Average document size 0.15 x x x = = bytes

13 Web server workload characterization Document request arrival rate is function of the think time (CLi, TR, Web) #Users i Response time + think time #Users i 48 per Lan 1 7 per Lan 2 15 per Lan 3 15 per Lan 4 45 sec

14 Web server workload characterization Device service time –CPU: 1 msec processing time x HTTP request –Disk: We need to consider Seekrand= avg time to position at a random cylinder DiskRevTIme = time for a complete disk revolution TransferTime = BlockSize/ 10 6 x TransferRate ControllerTime = time spent at the controller for an I/O req. S d = ControllerTime +Pmiss x (SeekRand + DiskRevolutionTime/2+TransferTime)

15 Web server workload characterization Lan hp: no fragmentation i.e. max data area 1500 bytes; hp no data overhead for HTTP request NDatagrams = ServiceTime n = MessageSize + TCPOvhd min n MTU n - IPOvhd Overhead n = TCPOvhd+Ndatagrams x (IPOvhd + FrameOvhd n ) 8 x (MessageSize + Overhead n ) 10 6 x Bandwidth

16 Web server workload characterization Lan hp: no fragmentation i.e. max data area 1500 bytes; hp no data overhead for HTTP request Ethernet NDatagrams = ServiceTime n = Overhead n = 20 + Ndatagrams x ( ) 8 x ( ) 10 6 x Bandwidth

17 Web server workload characterization Lan hp: no fragmentation i.e. max data area 1500 bytes; hp no data overhead for HTTP request Token ring NDatagrams = ServiceTime n = Overhead n = 20 + Ndatagrams x ( ) 8 x ( ) 10 6 x Bandwidth

18 Web server workload characterization Router delay 134 usec x packet (arrotondato a 1 msec totale) FDDI delay with ServiceTime n = 8 x (MessageSize + Overhead n ) 10 6 x Bandwidth

19 Local file system workload characterization File dimension 8192 bytes avg NFS request arrival rate is function of the think time (CLi, FS, FSi) #Users i Response time + think time #Users i 102 per Lan 1 43 per Lan 2 85 per Lan 3 85 per Lan 4 10 sec

20 Local file system workload characterization Device service time –CPU: 1 msec per file request –Disk: We need to consider Seekrand= avg time to position at a random cylinder DiskRevTIme = time for a complete disk revolution TransferTime = BlockSize/ 10 6 x TransferRate ControllerTime = time spent at the controller for an I/O req. N blocks to read = 8192/2048 = 4 –Lan i with 8192 bytes

21 Throughput & response time ClassThroughput Response time (req/sec)(sec) CL1, FS, FS1 10,12 0,08 CL2, FS, FS24,230,06 CL3, FS, FS38,440,08 CL4, FS, FS48,440,07 CL1, TR, WebS0,348,58 CL2, TR, WebS0,148,55 CL3, TR, WebS0,287,96 CL4, TR, WebS0,288,35