International Business Machines Corporation

Microsoft Technology Licensing, LLC

Sahil Suneja

Canturk Isci

Jim Alain Laredo

Yunhui Zheng

Alessandro Morari

Yufan Zhuang

Ramachandran Ramjee

Malgorzata Steinder

Salman Abdul Baset

Hai Huang

Shripad Nadgowda

Vasanth Bala

Todd W Mummert

Byung Chul Tak

Vishnu Navda

Jia Jun Brandon Lum

Ricardo Andrei Koller Jemio

Ananth Balashankar

Techniques regarding training one or more AI models for a source code understanding task are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a training component that can train an artificial intelligence model on source code samples for a source code understanding task. The source code samples can be ranked based on code complexity.

Complexity based artificial intelligence model training

Techniques regarding augmenting one or more training datasets for training one or more AI models are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise training augmentation component that can generate an augmented training dataset for training an artificial intelligence model by extracting a simplified source code sample from a source code sample comprised within a training dataset.

Training data augmentation via program simplification

Systems, computer-implemented methods, and computer program products that facilitate container inspection components of a container-based virtualization environment are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a container inspection control component that can define one or more constrained capabilities of a container inspection. The computer executable components can further comprise a container inspection component that can inspect a virtual container based on the one or more constrained capabilities.

Safe shell container facilitating inspection of a virtual container

Techniques facilitating security hardening systems that host containers are provided. In one example, a system comprises: a memory that stores computer executable components; and a processor that executes computer executable components stored in the memory. The computer executable components comprise: a boot component performs a portion of a trusted boot sequence to securely boot the system to a defined secure state wherein one or more types of administrative access to a container memory are deactivated. The computer executable components also comprise: a core service component started as a part of the trusted boot sequence and that securely obtains one or more decryption keys for use with the container memory; and a runtime decryption component that uses the one or more decryption keys to perform runtime decryption of one or more files accessed by a container associated with the container memory.

Blackbox security for containers

A sandbox component, operatively coupled to a host and a guest container, the sandbox component securely extends systems data collection software with potentially untrusted third-party code. A secure environment is enabled where plugins will run inside a sidecar container that is separate from a guest container. A container consists of an entire runtime environment: an application, plus its dependencies, libraries and other binaries, and configuration files needed to run it, bundled into one package. A sidecar service is not necessarily part of the application but is connected to the guest container and follows the parent application. A sidecar is independent from its primary application in terms of runtime environment and programming language. The sidecar plugin will be given a sparse/limited set of privileges required to simply perform its intended function and the Linux kernel constructs will control data access and transfer. The Plugin is essentially considered sandboxed as it runs in a sidecar and is fenced in by a set of kernel constructs.

Secure system state extraction software extensibility via plugin sandboxing

A mechanism is provided in a data processing system for monitoring a virtual machine in a distributed computing system. An out-of-virtual machine monitor receives, via an out-of-band path, disk and memory state data of the virtual machine exposed in a hypervisor hosting the virtual machine. The out-of-virtual machine monitor interprets logical information from the virtual machine disk and memory state data. The out-of-virtual machine monitor parses the logical information to extract related system or log information for a monitoring endpoint. The out-of-virtual machine monitor translates the system or log information to a standard format expected by the monitoring endpoint. The out-of-virtual machine monitor forwards a monitoring data stream to the monitoring endpoint.

Always-on monitoring in the cloud

In an approach for determining optimal performance-related configurations of applications by analyzing peer data, a processor monitors instances of an application, wherein the instances of the application are running on a plurality of devices. A processor receives data related to: configuration parameters of the application, system parameters of the plurality of devices, and performance metrics of the plurality of devices. A processor determines correlation and dependencies between the configuration parameters with associated system parameters, using: an all pair analysis and a Pearson product-moment correlation coefficient (PPMCC). A processor determines dependencies between the configuration parameters with performance metrics. A processor recommends to a user, configuration settings of the application, based on the analysis of the data.

Peer-based optimal performance configuration recommendation

Described herein are various technologies pertaining to scheduling data transfer between a mobile computing device and a base station in a cellular network. A signal quality value for a signal over which data is to be transferred is computed, and transfer of data is scheduled based upon the signal quality value. If the signal quality value is above a threshold, a wireless radio of the mobile computing device is caused to commence data transfer or continue data transfer. If the signal quality value is below the threshold, and the data need not be immediately transferred, then the wireless radio is caused to transition to an idle state or remain in an idle state.

Growing community of inventors

Yorktown Heights, NY, United States of America

Sahil Suneja