NetApp PowerPoint Template

Download Report

Transcript NetApp PowerPoint Template

AFS and NFS
… 20 Years Later
Mike Kazar
[email protected]
June 2005
Network Appliance
1
Overview

Inspired by talk by Brian Pawlowski (beepy)
–
–
presented to a bunch of folks at IBM
including lots of AFS people

What were AFS’s goals? NFS’s goals?

Innovation on client and server ends

Storage management
–
Then and now

Some thoughts about the future

Lessons learned

Questions
Network Appliance |
7/17/2015
2
Who Am I?
 Part
–
–
–
–
–
of AFS design team, with
Bob Sidebotham (Rx, Volumes)
Mike West (Server)
Sherri Nichols (VLDB &c)
M. Satyanarayanan (prototype RPC)
Dave Nichols (prototype client)
 Coded
–
–
cache manager
kernel port of Rx
 Now
Network Appliance |
parts of AFS
at NetApp
7/17/2015
3
AFS Goals in 1984
 Distributed
–
connect lots of workstations together
 Scalability
–
–
did I mention *lots*?
key approach was caching
•
the fastest RPC is the one you don’t make
 Security
–
needed isolation from random students
 Management
–
–
–
–
Network Appliance |
was an afterthought
but turned out to be critical
volumes
volume *moves*
mirroring
7/17/2015
4
Key points from Beepy’s talk
 What
–
–
–
–
 If
is NFS?
IETF standard
bundled with all Unix/Linux systems
available on nearly everything
heterogeneous (systems, FSes)
NFS is the answer, what was the question?
Network Appliance |
7/17/2015
5
Stick a fork in it…
FTAM
Network Appliance |
7/17/2015
6
Clients, Servers, an’ ‘at
 So,
why am I smiling?
 Clients,
–
Network Appliance |
Protocols, and Servers
Different goals for each
7/17/2015
7
Clients
 OS
integration
 Bug-free
–
or you get calls all the time
 Reference
–
–
–
Network Appliance |
implementation
helps portability a great deal
even though porting still lots of work
“Vnode” layers in Windows, AIX, Solaris, &c
7/17/2015
8
Protocols

Where NFS Really Shines
–
–
–

Public protocol spec
•
everyone knows they can implement it
Interoperability tests (Connectathon)
•
formally, helps certify who’s in the game
•
informally, helps communication!
–
set goals for future work
Reference implementation
•
as education tool
Where NFS sucks
–
–
Network Appliance |
cache coherence
•
how did they blow this in NFS version 4?
but still relatively minor
•
few applications use DSM
•
locking makes most things work
7/17/2015
9
Servers
 Where
–
–
–
Network Appliance |
AFS Shines
Data management
•
global name space
•
cell name spaces
•
transparent move
•
transparent load balancing mirrors
•
flexibly allocated volumes
•
snapshots / clones
Usable ACLs
•
suprisingly
caching
•
using memory caches today
7/17/2015
10
Today
 NFS
–
–
is the protocol of choice
open licensing made it a no-brainer
simplicity also a plus for growth
 Some
–
–
–
Snapshots on NetApp filers
flexible volumes
non-transparent moves
 Some
–
–
–
Network Appliance |
AFS data management available today
AFS data management available “soon”
transparent move
load balancing mirrors
multiple cells
7/17/2015
11
OnTAP NG Architecture
Q1
Location independence
acct random
Q2
P1
P2
Atomic move, snapshots
Class of service
Mirrors, caches
SpinServer
1
SpinServer
2
SpinServer
3
acct
Q1
Q2
P1
Unix
workstation
Network Appliance |
Windows
desktop
7/17/2015
Windows
desktop
random
P2
UNIX
workstation
12
Architecture Detail
ONTAP / NG:
2-Stage Distributed
File System
1.
2.
3.
Request switched to
appropriate back-end
IP-based cluster
network
Client Access
Client Access
Gigabit Ethernet
Gigabit Ethernet
Network Function
Network Function
• TCP termination
• VLDB lookup
• Protocol translation
to SpinFS
• TCP termination
• VLDB lookup
• Protocol translation
to SpinFS
SpinFS Protocol
X
No client code changes
Disk Function
• Caching
• Locking
Fibre Channel
Gigabit
Ethernet
Switch
Disk Function
• Caching
• Locking
Fibre Channel
12
Network Appliance |
7/17/2015
13
History as Knobs
 No
knobs
–
initial filers
•
one volume, period
Suns, &c
•
network parameters, exports, not much else
–
 Too
many knobs
–
–
–
 Just
–
–
Network Appliance |
tracking thousands of volumes
figuring out restores
where do I create a new volume?
right
self-managing based on guidance
eg. this part of the name space is a database
7/17/2015
14
The Future
 Policy-based
–
–
–
Network Appliance |
management
declare part of name space “database”
•
inherit RAID level, drive speed
•
referenced abstractly
constraint engine
•
moves data around
•
with limited system impact
•
when desired
tied to delegation
•
sub-admins tied to name space parts
•
sub-admins constrained by resource limits
7/17/2015
15
The Future
 Ties
–
–
–
–
Network Appliance |
to historical data
what volumes are heavily loaded, and when?
which volumes grow and shrink?
•
what variation in size
•
what variation in load
what volumes were on this server?
•
backup database issue
charge-back
7/17/2015
16
The Future
 Quality
–
–
–
Network Appliance |
of Service
Important for managing applications
but no common framework
•
eg: to connnect job controller and storage
so, need to start somewhere
•
virtual servers, perhaps
•
per volume?
•
priorities vs. guaranteed bandwidth or ops
7/17/2015
17
Lessons Learned
 Perfection
–
–
is highly overrated
POSIX semantics never really required
Huge effort in DCE/DFS
 Universities
–
–
–
–
–
Network Appliance |
are similar to enterprises
pointless politics and empire building
24x7 operation
availability and reliability
coordination with users nearly impossible
slightly cheaper
7/17/2015
18
Lessons Learned
 Technology
–
–
20 years to get AFS volume concept out
•
only happened because Blake Lewis and Ed.
Zayas went to NetApp
More for transparent moves
•
Spinnaker acquisition
 Beware
–
–
second system syndrome
and ignoring customer requirements
DCE/DFS was nearly complete waste of time
 Morris’s
–
–
Network Appliance |
Transfer is Hard
point about 3rd systems
independence from inventor is critical
•
throw out bad ideas
allows simplification
7/17/2015
19
Lessons Learned
 No
one pays attention to system management
–
but it is critical to any technology deployment
QoS is part of this
scaling is part of this (managing resource pools)
centralized error reporting
dynamic reconfiguration
–
–
–
–
 Don’t
–
–
–
Network Appliance |
be greedy!
know where your real value is
know how to get help
IBM lived in fear that Sun would productize AFS!
•
so bungled licensing
7/17/2015
20