Sam Idicula

Bio: Sam Idicula is an academic researcher from Business International Corporation. The author has contributed to research in topics: Efficient XML Interchange & XML database. The author has an hindex of 22, co-authored 86 publications receiving 1257 citations. Previous affiliations of Sam Idicula include Oracle Corporation.

Mechanism for uniform access control in a database system

Abstract: Techniques are provided for facilitating uniform access control to data managed by a database server that can emulate hierarchically organized systems, whether the data is accessed through hierarchical or relational access mechanisms. A database server that can emulate hierarchically organized systems uses separate relational or object-relational database tables to store the content of the resources that belong to a hierarchy (the “content structures”) and information that captures the hierarchy (the “hierarchy structures”). Both types of structures contain access control data that define consistent user access privileges. To determine access privileges for a user requesting access to data in the database, access control information is accessed in the hierarchy structures when the request is made through the hierarchical access mechanism, or accessed in the content structures when the request is made through a relational access mechanism. Access control is consistent between the hierarchical or relational access mechanisms because access through either is governed by user access data that reflects the same privileges.

Performing an action in response to a file system event

Abstract: A method and apparatus for performing an action in response to a file system event is provided. According to one aspect, sets of “event listeners” are associated with a file hierarchy and/or the nodes thereof. Each event listener contains a set of “event handlers.” Each event handler corresponds to a separate type of event that may occur relative to the file hierarchy's nodes. When an event is going to occur relative to the hierarchy or a node thereof, all event listeners that are associated with that hierarchy/node are inspected to determine whether those event listeners contain any event handlers that correspond to the event's type. Those event handlers that correspond to the event's type are placed in an ordered list of event handlers to be invoked. As the event handlers in the list are invoked, programmatic mechanisms that correspond to those event handlers are executed to perform customized user-specified actions.

Query optimization by specifying path-based predicate evaluation in a path-based query operator

Abstract: The approaches described herein provide an efficient way for a database server to process certain kinds of queries over XML data stored in an object-relational database that require the evaluation of a predicate expression with one or more path-based operands. A predicate expression part of a XQuery or SQL WHERE clause that returns a boolean value. A database server first determines whether the query qualifies for this particular kind of optimization, then rewrites the query using an enhanced query operator syntax for specifying the predicate expression to be evaluated. The enhanced query operator subsumes the work of a second path-based query operator, resulting in the suppression of the WHERE EXISTS subquery. The rewritten query operator is used to generate a query execution plan that provides for several query execution optimizations.

Flow-Join: Adaptive skew handling for distributed joins over high-speed networks

Abstract: Modern InfiniBand interconnects offer link speeds of several gigabytes per second and a remote direct memory access (RDMA) paradigm for zero-copy network communication. Both are crucial for parallel database systems to achieve scalable distributed query processing where adding a server to the cluster increases performance. However, the scalability of distributed joins is threatened by unexpected data characteristics: Skew can cause a severe load imbalance such that a single server has to process a much larger part of the input than its fair share and by this slows down the entire distributed query. We introduce Flow-Join, a novel distributed join algorithm that handles attribute value skew with minimal overhead. Flow-Join detects heavy hitters at runtime using small approximate histograms and adapts the redistribution scheme to resolve load imbalances before they impact the join performance. Previous approaches often involve expensive analysis phases, which slow down distributed join processing for non-skewed workloads. This is especially the case for modern high-speed interconnects, which are too fast to hide the extra computation. Other skew handling approaches require detailed statistics, which are often not available or overly inaccurate for intermediate results. In contrast, Flow-Join uses our novel lightweight skew handling scheme to execute at the full network speed of more than 6 GB/s for InfiniBand 4×FDR, joining a skewed input at 11.5 billion tuples/s with 32 servers. This is 6.8× faster than a standard distributed hash join using the same hardware. At the same time, Flow-Join does not compromise the join performance for non-skewed workloads.

Techniques for increasing efficiency while servicing requests for database services

Abstract: Techniques for servicing requests for database services include maintaining at a database server an available set of one or more database session data structures. Each database session data structure holds information to support one session of one or more requests for database services over a communication connection that persists for one or more communications from one client. A database session data structure in the available set is not associated with any client currently connected to the database server. These techniques allow a database server to more efficiently service more numerous requests for database services, such as generated by communications using a stateless protocol like HTTP.

Parallel convolutional processing using an integrated photonic tensor core.

Abstract: With the proliferation of ultrahigh-speed mobile networks and internet-connected devices, along with the rise of artificial intelligence (AI)1, the world is generating exponentially increasing amounts of data that need to be processed in a fast and efficient way. Highly parallelized, fast and scalable hardware is therefore becoming progressively more important2. Here we demonstrate a computationally specific integrated photonic hardware accelerator (tensor core) that is capable of operating at speeds of trillions of multiply-accumulate operations per second (1012 MAC operations per second or tera-MACs per second). The tensor core can be considered as the optical analogue of an application-specific integrated circuit (ASIC). It achieves parallelized photonic in-memory computing using phase-change-material memory arrays and photonic chip-based optical frequency combs (soliton microcombs3). The computation is reduced to measuring the optical transmission of reconfigurable and non-resonant passive components and can operate at a bandwidth exceeding 14 gigahertz, limited only by the speed of the modulators and photodetectors. Given recent advances in hybrid integration of soliton microcombs at microwave line rates3-5, ultralow-loss silicon nitride waveguides6,7, and high-speed on-chip detectors and modulators, our approach provides a path towards full complementary metal-oxide-semiconductor (CMOS) wafer-scale integration of the photonic tensor core. Although we focus on convolutional processing, more generally our results indicate the potential of integrated photonics for parallel, fast, and efficient computational hardware in data-heavy AI applications such as autonomous driving, live video processing, and next-generation cloud computing services.

Parallel convolution processing using an integrated photonic tensor core

Abstract: With the proliferation of ultra-high-speed mobile networks and internet-connected devices, along with the rise of artificial intelligence, the world is generating exponentially increasing amounts of data - data that needs to be processed in a fast, efficient and smart way. These developments are pushing the limits of existing computing paradigms, and highly parallelized, fast and scalable hardware concepts are becoming progressively more important. Here, we demonstrate a computational specific integrated photonic tensor core - the optical analog of an ASIC-capable of operating at Tera-Multiply-Accumulate per second (TMAC/s) speeds. The photonic core achieves parallelized photonic in-memory computing using phase-change memory arrays and photonic chip-based optical frequency combs (soliton microcombs). The computation is reduced to measuring the optical transmission of reconfigurable and non-resonant passive components and can operate at a bandwidth exceeding 14 GHz, limited only by the speed of the modulators and photodetectors. Given recent advances in hybrid integration of soliton microcombs at microwave line rates, ultra-low loss silicon nitride waveguides, and high speed on-chip detectors and modulators, our approach provides a path towards full CMOS wafer-scale integration of the photonic tensor core. While we focus on convolution processing, more generally our results indicate the major potential of integrated photonics for parallel, fast, and efficient computational hardware in demanding AI applications such as autonomous driving, live video processing, and next generation cloud computing services.

Enforcing network service level agreements in a network element

Abstract: Enforcing network service level agreements in a network infrastructure element comprises receiving, at the network infrastructure element, an application-layer message comprising one or more of the packets; forwarding the application-layer message toward a destination endpoint and concurrently copying the application-layer message without disrupting the forwarding; using the copied application-layer message, discovering one or more applications or services that are using the network; using the copied application-layer message, identifying one or more network-layer condition metrics, and identifying one or more application-layer condition metrics; determining, based on the identified network-layer condition metrics and the application-layer condition metrics, whether one or more conditions of a service level agreement are violated; and in response to determining a violation, performing one or more responsive operations on one or more network elements.