Breaking News
News & Analysis

Mellanox rolls 40G Ethernet switch

4/25/2011 12:30 PM EDT
1 Comment
NO RATINGS
More Related Links
View Comments: Newest First | Oldest First | Threaded View
docdivakar
User Rank
Author
re: Mellanox rolls 40G Ethernet switch
docdivakar   4/25/2011 9:31:37 PM
NO RATINGS
@Rick Merritt: it was good to see you at GreenNet last week in the City. Question: what is the Ethernet PMD for the switch? RJ-45? If that is the case, then your statement needs revision: "Switch X consumes about 2W per 56 Gbit/s Infiniband port, 1.5W on 40GE ports and 0.62W on 10GE ports. That's about on par with competing Ethernet devices some of which require external physical layer components that the Switch X integrates." Because the 10Gig RJ45 ports today consume more than 5W. If Switch X supports other (than RJ45) PMD's for Ethernet (like SFP+), then the power numbers look reasonable; I can also see how Mellanox leveraged its effort in IB switches to develop a product for 100G/40G Ethernet. MP Divakar

Most Recent Comments
michigan0
 
SteveHarris0
 
realjjj
 
SteveHarris0
 
SteveHarris0
 
VicVat
 
Les_Slater
 
SSDWEM
 
witeken
Most Recent Messages
9/25/2016
4:48:30 PM
michigan0 Sang Kim First, 28nm bulk is in volume manufacturing for several years by the major semiconductor companies but not 28nm FDSOI today yet. Why not? Simply because unlike 28nm bulk the LDD(Lightly Doped Drain) to minimize hot carrier generation can't be implemented in 28nm FDSOI. Furthermore, hot carrier reliability becomes worse with scaling, That is the major reason why 28nm FDSOI is not manufacturable today and will not be. Second, how can you suppress the leakage currents from such ultra short 7nm due to the short channel effects? How thin SOI thickness is required to prevent punch-through of un-dopped 7nm FDSOI? Possibly less than 4nm. Depositing such an ultra thin film less then 4nm filum uniformly and reliably over 12" wafers at the manufacturing line is extremely difficult or not even manufacturable. If not manufacturable, the 7nm FDSOI debate is over!Third, what happens when hot carriers are generated near the drain at normal operation of 7nm FDSOI? Electrons go to the positively biased drain with no harm but where the holes to go? The holes can't go to the substrate because of the thin BOX layer. Some holes may become trapped at the BOX layer causing Vt shift. However, the vast majority of holes drift through the the un-dopped SOI channel toward the N+Source,...

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed