The pressure is always a challenge in a XenApp / XenDesktop design, like so many other criteria in be eligible.
Note: I use now, but from a print design point of view, it is not much change between XenApp and XenDesktop, unless you are using RemotePC feature of XenDesktop (more on XenApp that at another time?)
on eDocs some scenarios are covered, and there is a question I am often asked by customers if it on a XenApp design work on the site the print server refers. We always say that the data on the XenApp server to be close. Is it the same for the print server location?
Various printing design possibilities
Often in XenApp architecture is the print server in the data center, near the XenApp server. With this construction, the important RAW bandwidth is sent by the print server to the actual printer. It is also worth using SMB protocol between XenApp notes and the print server:
If you add Citrix Universal Print Server component on the Configuration, HTTP SMB replace, but the critical part is the raw bandwidth to the remote printer:
If the print server (here with UPS component) in the branch office is the focus changes from RAW printer data to the traffic than the one to be sent over the wire between XenApp and the print server:
the goal is to determine today which printer driver language provides lower bandwidth usage where the location of a print server is best suited, based on the type of document and printing technology (Microsoft print Server, Citrix UPS, UPS + Citrix UPD) used.
course, only no one prints a single document type, however, based on statistics, you can get in your business, these test results can assist you in your decision.
Lab configuration
I was with Citrix UPS 7.6 (on WS08R2 SP1 compatible as WS2012R2 version is not yet publicly available) with latest hotfixes and XenApp 7.6 on WS2012R2 with latest hotfix (including UPS7.6 hotfixes) to connect to a published desktop. Everything is hosted on XenSerer. The printer is a Canon iRC2030, as IP printer on UPS VM with the UFRII, PS3, PCL 5e and PCL6 driver explained (all 14.02 release other than PS3, version 21.52)
. Note: UFRII proprietary page description language (PDL) from Canon is developed. I recommended, for more information on this technology Canon website
The tests
printer defined as a session printer performed in XenApp and 3 tests :. UPS disabled UPS with native drivers and UPS with UPD driver.
Example DOCX, XLSX, PPTX and PDF documents were used (Share File-Link), all of which come from the Citrix Web site. Microsoft Office 2010 and Adobe Acrobat Reader 10.1.4 used. Wireshark running on the UPS-VM, sent the amount of data to UPS and to the printer to collect.
The tests are carried out without UPS used as a reference, Bytes and method of transmission to the endpoint deployment (either UPS or printer). However, the duration of the printer is not comprise the actual printing time that the time for the entire print job is necessary to physically print. Why? because I was far away from the printer when I my test run ... 
Disclaimer: are the numbers not real life figures, but from a laboratory environment , collected with sample documents. You should perform similar tests to determine what is most appropriate in your area.
Microsoft Network Printing results
| DOCX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | 13.5 | 3,639,365 | 11.8 | 780 641 | 14 | 2471111 |
| PRN | 7.7 | 3,234,527 | 2.57 | 1,439,569 | 2:12 | 2522645 |
| PPTX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | 4.8 | 2,519,07 | 4.9 | 942 156 | 5.5 | 1,372,595 |
| PRN | 3.9 | 2,350,864 | 0.9 | 1,935,805 | 0.7 | 1,391,463 |
| PCL5e | PCL6 | UFRII | ||||
| time | byte | time | byte | time | byte | |
| PS | 9.2 | 1658351 | 7.3 | 478 591 | 8.6 | 1,150,946 |
| to PRN | 1.5 | 1374404 | 0.8 | 523 063 | 0.8 | 06 731 | [1945024[
Citrix UPS with native drivers
| DOCX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | 13.3 | 2,447,651 | 14 | 4,711,282 | 11.1 | 5,522,323 |
| PRN | 3.5 | 2,921,143 | 2.6 | 1,439,798 | 4.9 | 2,119,169 |
| XLSX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | [194502679]34560 | 12.5 | 3,784,939 | 14.3 | 6,702,008 | |
| to PRN | 2.1 | 2710400 | 1.8 | 482 031 | 2.0 | 1,511,389 |
| PPTX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | 6.2 | 1,611,616 | 6.3 | 2756111 | 6.8 | 2,304,349 |
| PRN | 2.1 | 3,330,415 | 0.9 | 1,884,01 | 0.7 | 1,391,469 |
| PCL5e | PCL6 | UFRII | ||||
| time | byte | time | byte | time | byte | |
| PS | 8.0 | 3,382,016 | 5.6 | 1,007,188 | 6.6 | 1664938 |
| PRN | 0.7 | 1,374,692 | 0.7 | 523237 | 0.7 | 06 461 |
Citrix UPS with Citrix UPD driver
| DOCX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | 21.6 | 7,372,622 | 16.8 | 5,819,772 | 36.7 | 14165652 |
| PRN | 11.5 | 3077676 | 9.9 | 1,332,327 | 14.2 | 2,523,394 |
| XLSX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | 7.9 | 3,456,0 | 12.5 | 3,784,939 | 14.3 | 6,702,008 |
| PRN | 2.1 | 2710400 | 1.8 | 482 031 | 2.0 | 1,511,389 |
| PPTX | PCL5e | PCL6 | UFRII | |||
| time | byte | time | byte | time | byte | |
| PS | 6.2 | 1,611,616 | 6.3 | 2,756,111 | 6.8 | 2,304,349 |
| PRN | 2.1 | 3,330,415 | 0.9 | 188401 | 0.7 | 1,391,469 |
| PCL5e | PCL6 | UFRII | ||||
| time | byte | time | byte | time | byte | |
| PS | 8.0 | 3,382,016 | 5.6 | 1,007,188 | 6.6 | 1,664,938 |
| to PRN | 0.7 | 1,374,692 | 0.7 | 523237 | 0.7 | 06 461 [1945022[ |
XenApp server traffic to print
to facilitate reading, a graphical view be easiest to transport XenApp Server shows to print, depending on the type of printer used (PCL5, PCL6 or UFRII) and printing technology.
With a PCL6 driver seems the most logical solution to be UFRII compared to PCL5 and. It is worth mentioning, the overhead generated by UPS via native Microsoft pressure, and I would recommend if you want to evaluate UPS to test with native drivers. Why? Since UPS embeds print traffic in XTE Service (Apache based service), which can lead NetScaler TCP / HTTP optimization and load balancing (which is another topic about a day).
to facilitate reading
Print Server printer traffic
, a graphical representation of the easiest, the movement of the print server to mourn, printer shows on the type of printer used depends (PCL5, PCL6 or UFRII) and printing technology. 
PCL6 is the best choice (I can not explain the results with PDF and UPD when on PCL6 printing, even if multiple tests) when the bandwidth optimization between the print server and printer is the ultimate goal.
Summary
were carried out from these tests, UPS will not help to reduce the bandwidth, but also offers a free universal printing solution (more no drivers on XenApp! ) and may benefit from NetScaler load balancing and TCP optimization (confirmed in another blog post ...). PCL6 seems that a suitable choice for everyone in order to limit the pressure range to either the print server or to the printer.
0 Komentar