Two-pronged strategy sparks open source initiative for cloud instrumentation and a highly distributed cloud-enabled monitoring platform</p>
SAN FRANCISCO, Mar 03, 2009 – CITTIO®, the innovation leader in automated network and systems monitoring software, lays out its vision and roadmap for cloud monitoring today in order to accelerate its adoption by enterprises. CITTIO's vision is to provide enterprise application owners as well as cloud operators with the tools they need for selecting, benchmarking and monitoring the most eligible applications and systems for cloud computing. To that end, CITTIO has embarked on a two-pronged strategy. First the company is leading the development of deep instrumentation for the vast array of cloud components through the power of the open source community. Second it is developing a highly distributed version of its well established enterprise management platform, WatchTower, to provide a seamless view of infrastructure and application health and audited service levels across the internal datacenter and remote cloud environments.... read more
All classes and variables accesible with the Hadoop providers are described in the HadoopProvider.doc document which is part of the distribution.
This release includes new Hadoop providers for NameNode, DataNode and JobTracker. All classes and variables accesible with these providers are described in the HadoopProvider.doc document
First Haddop HDFS provider is included with this release. The accesable HDFS parameters and statistics are:
CapacityTotal - total raw bytes including non-dfs used space.
CapacityUsed - total used space by data nodes
CapacityRemaining - total none-used raw bytes
TotalLoad - total number of active connections to the data nodes
BlocksTotal - total number of blocks in the system
FilesTotal - number of files in the system
PendingReplicationBlocks - number of blocks that are getting replicated at this instant.
UnderReplicatedBlocks - number of blocks that are waiting (queued) to be replicated at this instant.
ScheduledReplicationBlocks - number of blocks scheduled for replication
FSState - state of the file system. Valid values are "safeMode” or “Operational”
DeadDataNodes - number of Dead data nodes on the system
LiveDataNodes - number of Live data nodes on the system... read more
Zeppelin 0.1.3 alpha release is available for download. It includes 5 new benchmark providers:
Lmbench Service Provider user guide document is available for review
Zeppelin 0.1.2 alpha release is available for download. It includes support for some of the benchmarks included in Lmbench-3.0-a9
1. lat_fs - measure file system create/delete performance
2. lat_pipe - measure interprocess communication latency through pipes
3. lat_proc - process creation tests
4. lat_syscall - time simple entry into the operating system
5. mhz - calulate processor clock rate
This position paper outlines a technical architecture for effective Cloud monitoring and management. Project Zeppelin is a key part of the architecture as it makes advanced cloud instrumentation freely available to the IT management industry. The proposed component stack is drawn from open standards like WBEM/CIM XML and WS-Man, uses instrumentation from well known open source projects and leverages CITTIO's considerable experience in the Enterprise Network and Systems management area. ... read more
Zeppelin 0.1.1 includes support for XenServer 5.0. The new Zeppelin alpha release is available for download.
The initial alpha release of project Zeppelin is available for download. Only support for CentOS 5.2 is provided with this release.