Enterprise OpenStack projects confront underline gaps
Sunday, October 21st, 2012SAN DIEGO — For enterprises kicking a tires on a OpenStack cloud government platform, there
are still some pivotal mixture blank from a stream release.
These gaps embody monitoring, high accessibility (HA), business smoothness (BC) and integration
with existent user government systems, as good as issues with ascent support and documentation,
according to attendees and presentations during this week’s OpenStack Summit.
Monitoring is a pivotal pain indicate for James Penick, an designer during Yahoo Inc., that is looking to
deploy tens of thousands of nodes on OpenStack by 2014. “There are solutions out there we can pay
for,” he said, “but that’s a outrageous opening in OpenStack right now.”
Storage monitoring that’s adult to standard with existent craving collection is also a blank element,
according to Pete Johnson, an operative during Hewlett-Packard Co. (HP). For example, if users today
want to get a normal age or distance of objects in a Swift storage container, that information has
to be collected by particular queries; that routine needs to be lifted to a turn of existing
storage government collection used in enterprises, Johnson said.
Enterprises, however, shouldn’t indispensably reason their exhale watchful for low-level monitoring
to turn partial of core OpenStack.
“It doesn’t unequivocally make clarity for us to build it,” pronounced Josh McKenty, OpenStack Foundation board
member and CEO during Piston Cloud Computing Inc. “But we’re also hard-pressed to collect that [existing
open source monitoring tool] to include.”
In his prior purpose during NASA, McKenty’s group used mixed open source monitoring collection –
namely, Nagios, Munin and Ganglia — to guard opposite elements of OpenStack.
HA and DR: OpenStack prohibited potatoes
High accessibility and BC are dual some-more areas of discuss as to who should build that facilities for
OpenStack. Right now, users are speedy to confederate existent open source tools, such as a HA
Pacemaker application or DRBD [Distributed Replicated Block Device], an HA clustered storage file
system.
But not everybody trusts these tools. “There are lots and lots of open comments [on them] and
things [about them] that can break,” Yahoo’s Penick said.
While HA competence be a no-brainer for enterprises, it competence not be such a healthy fit for
OpenStack, that is focused on newer cloud-based applications that pattern for disaster rather than
requiring HA from a underlying infrastructure. Enterprises are also awaiting formation into
existing environments for user management, according to HP’s Johnson, that means formation with
Active Directory and Lightweight Directory Access Protocol (LDAP).
There are 4 packages for Active Directory formation now submitted to OpenStack for
future releases.
Database changes and support shortcomings
The rate of change in OpenStack in a dual years it’s been grown has been wily to keep up
with, Penick said. For example, there have been changes to a Nova database schema between the
Essex and Folsom releases that streamline queries though also impede upgrades.
“I positively don’t remonstrate with changing [database] schema between revisions,” he said. But
converting between a Essex schema and a Folsom schema is easier pronounced than done.
Currently, there are no collection to do this conversion.
“OpenStack cloud operators have to write a collection required to make this happen, do it by hand,
or do zero during all,” Penick said. “I spend a good volume of time vocalization with other OpenStack
operators, listening to their stories,” he said. “More than a few have answered a ‘How do you
upgrade?’ doubt by saying, ‘We don’t. Build a new cluster. Tell everybody to reconstruct their
VMs.’”
Another area of change between Essex and Folsom that’s tripped adult some users has been the
extraction of retard storage, formerly famous as Nova Volumes, into a apart application
programming interface called Cinder.
“It used to be that information about volumes and instances could always be found in a same
database,” McKenty explained. Some users “cheated,” in McKenty’s words, formulating certain method
calls around that assumption, that no longer applies.
Finally, users are job for improved high-level documentation, that would be particularly
helpful in troubleshooting problems for infrastructure operations teams, pronounced one developer working
as an eccentric executive with a vital European telecom. “If there’s something wrong, we still
have to find what’s wrong,” he said. “It’s easy for me as a developer to demeanour during a code, though it’s
not that easy for infrastructure operations people.”
The OpenStack Foundation skeleton to tell support of this inlet soon, McKenty said.
Piston Cloud also offers an “OpenStack 101″ white paper.
Beth Pariseau is a comparison news author for SearchCloudComputing.com and
SearchServerVirtualization.com. Write to her during bpariseau@techtarget.com or follow @PariseauTT on
Twitter.
Article source: http://www.pheedcontent.com/click.phdo?i=0465b931f40e8aa423fc00070a2d7259