Abstract: Systems and methods for identifying attack patterns or suspicious activity can include a profile builder, a primitive creator, and a compromise detector. The profile builder can populate one or more baseline activity profiles for each client of the plurality of clients or entities associated therewith. The primitive creator can create primitives by comparing identified or extracted features to information in the one or more baseline activity profiles. The compromise detector can receive primitives, and based on identified combinations or sequences of primitives, generate compromise events to be provided to clients.

Abstract: Technologies for dynamic loading of integrity protected modules into a secure enclave include a computing device having a processor with secure enclave support. The computing device divides an executable image into multiple chunks, hashes each of the chunks with corresponding attributes that affect security to generate a corresponding hash value, and generates a hash tree as a function of the hash values. The computing device generates an initial secure enclave memory image that includes the root value of the hash tree. At runtime, the computing device accesses a chunk of the executable image from within the secure enclave, which generates a page fault. In response to the page fault, the secure enclave verifies the associated chunk based on the hash tree and accepts the chunk into the secure enclave in response to successful verification. The root value of the hash tree is integrity-protected. Other embodiments are described and claimed.

Abstract: A secure storage module of a client device interacts with a set of secure storage servers to securely store data items of the client on the servers, such that no individual server has the data in readable (non-obfuscated) form. Additionally, the client secure storage module and the servers interact to allow the client device to read a given portion of the original data items from the servers, such that none of the servers can determine which portion of the original data is being requested. Similarly, the interactions of the client secure storage module and the servers allows the client device to update a given portion of the original data on the servers to a new value, such that none of the servers can determine which portion is being updated and that none of the servers can determine either the prior value or new value or the difference between the new value and the prior value.

Abstract: A technique for the secure exchange and verification of data between multiple entities that use compatible business management systems utilizes serializable data transfer objects to transfer business data over a secure communication path. A conversion algorithm transforms a subset of the business data contained within the transfer objects, and the resulting data string is compared to data strings from other objects that are related to the transaction and which were transformed by the same conversion algorithm. If the data strings are not identical, a discrepancy exists. Serializable objects link all data objects relating to a given business transaction. An interface allows users to view and manage data contained in data objects, including discrepancies between compared objects.