International Business Machines Corporation

Suriya Subramanian

Animesh Nandi

Prasad Manikarao Deshpande

Anindya Neogi

Sriram Raghavan

Gargi Banerjee Dasgupta

Atri Mandal

Narendran Sachindran

Methods and arrangements for efficiently mining a control flow graph from execution logs of a distributed system. Using at least one text clustering technique, two text clusters are generated from the plurality of execution logs. At least one approximate template is generated based on the at least two text clusters. At least one refined template is created via refining the at least one approximate template using multimodal sequencing. The control flow graph is created based on the at least one refined template. An anomaly is detected in the control flow graph.

Scalable and accurate mining of control flow from execution logs across distributed systems

Methods, systems, and computer program products for organizing on-disk layout of index structures are provided herein. A method includes organizing an on-disk corpus of data index structures; partitioning each data index structure into an active and/or latest state data partition and an archived data partition; sub-partitioning each of the archived data partitions into sub-partitions based on exact lifespan of each data item; executing a search query that includes temporal predicates and keyword predicates across the corpus by: pruning a collection of identified items of data from the corpus that contain the keyword predicates using range metadata associated with the temporal predicates to generate a pruned list of data items; and performing a list intersection operation on the data index structures to generate a list of common data items from the pruned list; retrieving the common data items from the active and/or latest state data partitions and the archived data sub-partitions.

Organizing on-disk layout of index structures to support historical keyword search queries over temporally evolving data

Methods and arrangements for identifying root causes of system failures in a distributed system said method including: utilizing at least one processor to execute computer code that performs the steps of: recording, in a storage device, collected machine state data, wherein the collected machine state data are added to historical machine state data; creating, based on the historical machine state data, a healthy map model; detecting at least one failed machine state in the distributed system; comparing the failed machine state against the healthy map model; identifying, based on the comparison, at least one root cause of the failed machine state; and displaying, on a display device, a ranked list comprising the at least one root cause. Other variants and embodiments are broadly contemplated herein.

Identifying root causes of failures in a deployed distributed application using historical fine grained machine state data

Growing community of inventors

Chennai, India

Suriya Subramanian