Abstract: A distributed computing system uses dynamically calculates a subset size for each of a plurality of load balancers. Each of a plurality of load balancers logs requests from client devices for connections to back-end servers and periodically sends a request report to a traffic aggregator, which aggregates the report requests from the load balancers in the corresponding zone. Each traffic aggregator sends the aggregated request data to a traffic controller, which aggregates the request data to determine a total number of requests received at the system. The total request data is transmitted through each traffic aggregator to each load balancer instance, which calculates a percentage of the total number of requests produced by the load balancer and determines a subset size based on the calculated percentage.

Abstract: Described herein is a computer-implemented method. The method comprises receiving an operation notification in respect of a gated operation from a change requesting system, determining an issue type associated with the gated operation, and creating an issue of the determined issue type. The method further comprises determining that the issue has transitioned state from a pending workflow state to a particular operation resolution workflow state and, in response, generating an operation resolution message which is communicated to the change requesting system.

Abstract: Systems and methods are disclosed for switching between batch processing and real-time processing of time series data, with a system being configured to switch between a batch processing module and a real-time processing module to process time series data. The system includes an orchestration service to indicate when to switch, which may be based on a switching event identified by the orchestration service. In some implementations, the orchestration service identifies a switching event in incoming time series data to be processed. When a batch processing module is to be used to batch process time series data, the real-time processing module may be disabled, with the real-time processing module being enabled when it is used to process the time series data. In some implementations, the real-time processing module includes the same processing models as the batch processing module such that the two modules' outputs have a similar accuracy.

Abstract: In some implementations, a device may determine whether an existing record for a file, received at a file storage location, is present in a record log. The device may generate, based on determining that the existing record for the file is not present in the record log, a record for the file in the record log. The record may identify the file and a first region. The device may determine, after generating the record, whether the record has been changed to identify a second region. The device may process the file based on determining that the record has not been changed to identify the second region.

Abstract: Technologies are described for managing shared read-only objects in hybrid distributed database systems. For example, data structures can be maintained for managing the shared read-only objects, comprising a data structure storing a “deleted at” value for each shared read-only object, a data structure storing a “deleted at” value for each secondary node, a data structure storing a indication of which secondary nodes are reading which shared read-only objects, and a data structure storing a reference count indicating a number of active read transactions for each shared read-only object. The shared read-only objects that can be garbage collected can be identified using, at least in part, the data structures. The identified shared read-only objects can then be garbage collected.

Abstract: Certain aspects of the present disclosure provide techniques for efficient data parity. Embodiments include receiving, by a first data consuming component, from a publication service, a plurality of database change records indicating changes to an underlying data source. Embodiments include updating, by the first data consuming component, a secondary data store based on the plurality of database change records. Embodiments include selecting a sample subset of the plurality of database change records based on one or more conditions. Embodiments include, for each respective database change record in the sample subset that has already been consumed by a second data consuming component that updated a primary data store based on one or more database change records: comparing the respective database change record to first data from the primary data store that corresponds to the respective database change record and making a parity determination based on the comparing.

Abstract: The disclosed embodiments provide a distributed transaction system including a group of validator nodes that are known to each other in a network but are indistinguishable to other network nodes. The validator nodes form a Committee including a Leader node and one or more Associate nodes configured to receive and process transaction requests and candidate requests, for example, to add new blocks to one or more blockchains. The Committee may be dynamically changed, such that new network nodes may be added to the Committee or may replace existing validator nodes. The Associate nodes also may coordinate with each other to select a new Leader node. The system may allow multiple request-fulfillment process to run simultaneously, thereby enhance the efficiency of the system.

Abstract: A clustered pair of storage systems configured for active-active bidirectional synchronous replication expose a stretched volume over paths to both storage systems. Writes to the stretched volume received at each system are replicated to the peer system. The cluster can use a time-to-live (TTL) mechanism by which a non-preferred system continuously requests a TTL grant from the preferred system to remain in the cluster. Algorithms that reduce or avoid data unavailability are described and can include assessing the health of the systems in the cluster. An unhealthy system can trigger a one-sided polarization algorithm to notify the peer system that it is polarization winner. An improved polarization technique using a witness to decide the polarization winner includes a system adding a time delay before contacting the witness if the system is unhealthy. A control component can detect an unhealthy system and disable the active-active bidirectional synchronous replication.

Abstract: An approach is disclosed for optimizing jobs, which involve effective data retrieval across multiple data sources, via an externalized query pattern. The approach receives a query. The approach allocates computing power to one or more computing nodes to process a job for the received query. The approach determines whether the job for the received query corresponds to an identifier indexed in a search engine. The approach retrieves, in response to determining the job for the received query corresponds to the indexed identifier, attribute data corresponding to the indexed identifier from the search engine. The approach provides a response to the received query request, in which the response includes the attribute data corresponding to the indexed identifier.

Abstract: Prior to initiating discovery of computing resources of an information technology (IT) infrastructure, an input query and a discovery job are received, and one or more trigger configuration items (CIs) are calculated. The input query specifies how the computing resources are discovered, and has a filter parameter. The discovery job is for the input query, has a job parameter, and is associated with a trigger query that is a subset of the input query. The trigger CIs are calculated based on the trigger query, such that the filter parameter of the input query is replaced with the job parameter in the trigger CIs. Discovery of the computing resources is initiated by dispatching each trigger CI. As discovery of the computing resources occurs, CIs representing the computing resources matching the trigger CIs are received, and the CIs are populated within a configuration management database (CMDB).

Abstract: Embodiments of the present disclosure provide methods, apparatus, systems, computer program products for providing a virtual network within an interactive electronic technical manual system (IETM) configured to provide electronic and credentialed access via an IETM viewer to technical documentation for an item. In one embodiment, a method is provided comprising: receiving data from a graphical code as a result of a user scanning the code; and responsive to receiving the data: identifying virtual networks supported within the IETM that are available for the user to join and are associated with a specific object of the item; providing an identifier for each of the virtual networks; receiving selection of an identifier for a particular virtual network; and responsive to receiving the selection, joining the user to the virtual network so that content that is specific to the object associated with the virtual network is available to the user via the IETM.

Abstract: A method includes monitoring a job being executed at the source mainframe. A job comprises multiple tasks. A method includes monitoring a particular task of the multiple tasks being executed at a source mainframe and determining an application required to execute the particular task. In response to determining that the particular task requires an application to execute, determining a target mainframe where the application is installed. A method further includes validating the environment of the target mainframe to confirm that the particular task can be executed using the target mainframe, and upon validating the target mainframe, redirecting the particular task to the target mainframe for execution. A method also includes monitoring the particular task being executed at the target mainframe and returning the results of the particular task from the target mainframe to the source mainframe.

Abstract: Aspects of the present invention disclose a method, computer program product, and system for auto-scaling a query engine. The method includes one or more processors monitoring query traffic at the query engine. The method further includes one or more processors classifying queries by a plurality of service classes based on a level of complexity of a query. The method further includes one or more processors comparing query traffic for each service class with a concurrency threshold of a maximum number of queries of the service class allowed to be concurrently processed. The method further includes one or more processors instructing auto-scaling of a cluster of worker nodes to change a number of worker nodes available in the cluster based on the comparison, over a defined period of time, of the query traffic relative to a defined upscaling threshold and a defined downscaling threshold.

Abstract: A system, computer-implemented method, and/or non-transitory computer-readable media may provide a warm start technique for cloud-hosted functions. Included is a non-transitory memory storage comprising instructions, and one or more processors in communication with the memory, wherein the one or more processors execute the instructions to identify a cloud-hosted function configured to be executed within a cloud computing environment. The cloud-hosted function is instantiated within the cloud-hosted computing environment.

Abstract: Embodiments of the present disclosure provide methods, apparatus, systems, computer program products for performing functionality requiring network connectivity in an interactive electronic technical manual system (IETM) configured to provide electronic and credentialed access to technical documentation for an item. In one embodiment, a method is provided comprising: receiving input comprising informational data; and responsive to receiving the data: generating a graphical security code; receiving input indicating a scanning of the code, wherein the scanning of the code (1) is performed by a user using a remote device connected to a network and (2) causes access to scanning functionality provided by an application installed on the device; and responsive to receiving the input: generating a graphical form code comprising the data; and causing display of a third window comprising the code, wherein the code can be scanned using the application to send the data to a third party over the network.

Abstract: Techniques for using tree data structures to maintain a transactionally consistent set with support for time-travel queries are described. When a transaction commits, a new version of the tree data structure is created using a copy-on-write based method such that the tree shares internal nodes with previous trees to save space. This approach may be used in the implementation of a transactional data catalog in which the files that make up a table are stored in a transactional set.

Abstract: Concepts are presented which achieve a more efficient coding of coefficients of a transform block by use of dependent quantization and context adaptive entropy coding or achieve a coding of coefficients of a transform block in a manner which allows a more efficient coding even if a usage of dependent quantization is combined with the usage of context adaptive entropy coding.

Abstract: In an edge computing system deployment, a system includes memory and processing circuitry coupled to the memory. The processing circuitry is configured to obtain a workflow execution plan that includes workload metadata defining a plurality of workloads associated with a plurality of edge service instances executing respectively on one or more edge computing devices. The workload metadata is translated to obtain workload configuration information for the plurality of workloads. The workload configuration information identifies a plurality of memory access configurations and service authorizations identifying at least one edge service instance authorized to access one or more of the memory access configurations. The memory is partitioned into a plurality of shared memory regions using the memory access configurations. A memory access request for accessing one of the shared memory regions is processed based on the service authorizations.

Abstract: Techniques for processing event streams are provided that use streaming analytics to estimate a target user for dynamic assets. An exemplary technique includes receiving a batch of events that have position data for a potential target, determining one or more polygons that comprise the potential target based on the position data for the potential target, determining an approximate target viewership for the one or more polygons based on a profile or taxonomy of the potential target, associating an advertisement with the one or more polygons based on the approximate target viewership, receiving a continuous stream of events that have position data for a dynamic asset, determining that the one or more polygons include the dynamic asset based on the position data for the dynamic asset, obtaining the advertisement associated with the one or more polygons, and providing the advertisement to a sink for display on the dynamic asset.

Abstract: Embodiments herein describe computer-implemented methods, computer program products, and computer systems for updating data templates. In an embodiment, one or more processors may be configured for receiving change data from tenant spaces configured to update tenant data templates of the tenant spaces; generating metadata for each of the tenant spaces corresponding to the change data; determining least common change data based on the metadata; generating change request data for a multi-cloud management platform (MCMP) data template based on the least common change data; and transmitting the change request data to the one or more tenant data templates. The method may further include initializing the MCMP configured to execute a multi-tenant deployment model having the MCMP data template, wherein the MCMP is connected to the tenant spaces.

Abstract: A method, computer system, and a computer program product for objective-based compression level change is provided. The present invention may include storing a volume in a storage device, wherein the stored volume is compressed using an initial compression level. The present invention may also include checking a last access time of the stored volume in the storage device at a regular interval. The present invention may further include, in response to determining, based on the checked last access time, that the stored volume is not accessed at the regular interval, recompressing the stored volume in the storage device using a higher compression level, wherein the higher compression level includes a higher compression ratio than a compression ratio associated with the initial compression level.

Abstract: Methods, systems, and computer programs are presented for processing Ethernet packets at a Field Programmable Gate Array (FPGA). One programmable integrated circuit includes: an internal network on chip (iNOC) comprising rows and columns; clusters, coupled to the iNOC, comprising a network access point (NAP) and programmable logic; and an Ethernet controller coupled to the iNOC. When the controller operates in packet mode, each complete inbound Ethernet packet is sent from the controller to one of the NAPs via the iNOC, where two or more NAPs are configurable to receive the complete inbound Ethernet packets from the controller. The controller is configurable to operate in quad segment interface (QSI) mode where each complete inbound Ethernet packet is broken into segments, which are sent from the controller to different NAPs via the iNOC, where two or more NAPs are configurable to receive the complete inbound Ethernet packets from the controller.

Abstract: A scheduling application is used to acquire into memory of a first processing system running on a computer system, a first jobs load for execution by the first processing system, the first jobs load stored in a database's table and having a first load size, the database associated with the computer system. The scheduling application is used to acquire into memory of a second processing system, a second jobs load for execution by a second processing system. The second jobs load being included in the table and having a second load size being balanced with the first load size and including jobs not in the first jobs load. In the first processing system's memory, jobs in the first jobs load are scheduled for execution by the first processing system. In the second processing system's memory, jobs in the second jobs load are scheduled for execution by the second processing system.

Abstract: One embodiment provides a method, including: receiving at least one deep learning job for scheduling and running on a distributed system comprising a plurality of nodes; receiving a batch size range indicating a minimum batch size and a maximum batch size that can be utilized for running the at least one deep learning job; determining a plurality of runtime estimations for running the at least one deep learning job; creating a list of optimal combinations of (i) batch sizes and (ii) numbers of the plurality of nodes for running both (a) the at least one deep learning job and (b) current deep learning jobs; and scheduling the at least one deep-learning job at the distributed system, responsive to identifying, by utilizing the list, that the distributed system has necessary processing resources for running both (iii) the at least one deep learning job and (iv) the current deep learning jobs.

Abstract: A method for processing data exactly once using transactional stream writes includes receiving, from a client, a batch of data blocks for storage on memory hardware in communication with the data processing hardware. The batch of data blocks is associated with a corresponding sequence number and represents a number of rows of a table stored on the memory hardware. The method also includes partitioning the batch of data blocks into a plurality of sub-batches of data blocks. For each sub-batch of data blocks, the method further includes assigning the sub-batch of data blocks to a buffered stream; writing, using the assigned buffered stream, the sub-batch of data blocks to the memory hardware; updating a storage log with an intent to commit the sub-batch of data blocks using the assigned buffered stream; and committing the sub-batch of data blocks to the memory hardware.

Abstract: A virtual machine management service obtains a request to instantiate a virtual machine image (VMI) to implement a virtual network function (VNF). The request specifies a set of processor requirements corresponding to instantiation of the VMI. In response to the request, the service identifies, from a server comprising a set of processor cores, available processor capacity. The service determines, based on the available processor capacity and the set of processor requirements, whether to instantiate the VMI on to a subset of processor cores of the server. Based on this determination, the service instantiates the VMI on to the subset of processor cores to implement the VNF.

Abstract: A securities trading apparatus includes a memory; and a processor coupled to the memory, the processor configured to: extract transactions from trading transactions stored in the memory, the extracted transactions being one of order transactions or insert transactions, each of the order transactions being a buy order transaction or a sell order transaction; and generate a single trading transaction from the extracted transactions.

Abstract: A set of transaction handling computing elements comprise a network core that receive and process transaction requests into an append-only immutable chain of data blocks, wherein a data block is a collection of transactions, and wherein an Unspent Transaction Output (UTXO) data structure supporting the immutable chain of data blocks is an output from a finalized transaction. Typically, the UTXO data structure consists essentially of an address and a value. In this approach, at least one UTXO data structure is configured to include information either in addition to or in lieu of the address and value, thereby defining a Transaction Output (TXO). A TXO may have a variety of types, and one type includes an attribute that encodes data. In response to receipt of a request to process a transaction, the set of transaction handling computing elements are executed to process the transaction into a block using at least the information in the TXO.

Abstract: The disclosed embodiments provide a distributed transaction system including a group of validator nodes that are known to each other in a network but are indistinguishable to other network nodes. The validator nodes form a Committee including a Leader node and one or more Associate nodes configured to receive and process transaction requests and candidate requests, for example, to add new blocks to one or more blockchains. The Committee may be dynamically changed, such that new network nodes may be added to the Committee or may replace existing validator nodes. The Associate nodes also may coordinate with each other to select a new Leader node. The system may allow multiple request-fulfillment process to run simultaneously, thereby enhance the efficiency of the system.

Abstract: A computer system is provided for retrospectively processing a data structure that includes a plurality of entries. The computer system determines if certain data transactions requests that have been recorded in the data structure could have been executed differently (e.g., by being processed at an earlier point in time). For a given entry in the data structure, the system determines if data transaction request could have at least partly succeeded against a prior recorded state of at least one of two ordered lists of pending data transaction requests. Another entry is then found that caused the initial entry to fail in execution and a time delta is stored between the timestamp of the another entry and the timestamp of the initial entry.

Abstract: Concurrent threads may be synchronized at the level of the memory words they access rather than at the level of the lock that protects the execution of critical sections. Each lock may be associated with an array of flags and each flag may indicate ownership of certain memory words. A pessimistic thread may set flags corresponding to memory words it is accessing in the critical section, while an optimistic thread may read the corresponding flag before any memory access to ensure that the flag is not set and that therefore the associated memory word is not being accessed by the other thread. Thus, optimistic threads that do not have conflicts with the pessimistic thread may not have to wait for the pessimistic thread to release the lock before proceeding.

Abstract: At a resource corresponding to one node of a data processing workflow, a persistent representation of a binary signal representing a condition associated with an event is generated. The signal indicates a validity period of the condition. At a second resource corresponding to another node of the workflow, a decision to initiate an action is generated based on analysis of a set of binary signals and an action scheduling rule, and the action is initiated.

Abstract: Various techniques for facilitating communication with and across a clinical environment and a cloud environment are described. For example, a method for providing messaging in a clinical environment during a network outage is described. During a network outage, messages in the queue may become stale. A connectivity adapter may remove those messages from the queue. When the network connection is restored, the connectivity adapter may transmit any remaining messages in the queue.

Abstract: Various techniques for facilitating communication with and across a clinical environment and a cloud environment are described. For example, a method for providing messaging in a clinical environment during a temporary network interruption is described. A connectivity adapter may keep the messages in the queue despite the temporary network interruption. As long as the network connection is restored before the messages become stale, the connectivity adapter may be able to transmit the messages in the queue despite the network interruption.

Abstract: A method and system for reducing triggering of throughput penalties imposed on a group of users by a software-as-a-service (SaaS) server due to Application Programming Interface (API) calls exceeding limits of the SaaS server is disclosed. The approaches include actions of intercepting requests to the SaaS server from a user group and monitoring both a rate of API calls for the requests and a rate of API events generated by forwarding the API calls to the SaaS server, intercepting the SaaS server's responses, where some of the responses indicate a throughput penalty imposed by the server, inferring load conditions of the SaaS server by analyzing the varying rate of API events against the responses with imposition of throughput penalty and setting an API call throttle limit dynamically adaptive to the inferred load conditions, then throttling the rate of the API calls for the group's requests according to the throttle limit.

Abstract: An information processing system manages data on customers using tenants which are storage regions dedicated for the customers of a database and registers a message for processing data of a processing target in a queue. The data is processed by accessing the tenants corresponding to the data so as to refer to the tenants based on the message. The information processing system registers a message for processing a plurality of data of processing targets as a batch process in the queue in a case where the plurality of data satisfy the predetermined condition when the message for processing a plurality of data of processing targets is to be registered in the queue. In the process performed on the plurality of data, the different tenants are accessed to be referred to.

Abstract: Batch building for artificial intelligence workflows, including: issuing, responsive to a request for a batch of data objects, requests to a data repository for multiple data objects stored among one or more directories; selecting, in accordance with a batch building policy, a subset of data objects based on one or more responses to the requests; and providing, to the artificial intelligence workflow, a batch of data objects that includes the subset of data objects.

Abstract: A first user request which specifies a target document set wherein a first subset of the documents is flagged by a user. A primary flag table is created for the target document set. A first document subset is created matching the first user request. It is determined whether a number of flagged documents exceeds a first threshold. If so, a secondary flag table is created for the first document subset and flag data corresponding to the first document subset is stored in the secondary flag table. The flag data in the secondary flag table is merged into the primary flag table.

Abstract: Systems and methods for identifying a characteristic of an input code by converting the input code into simplified code and using the simplified code to generate snippets that can be compared to code in a database. Preferably, code is simplified by at least one of: unifying of capitalization, removing characters, and replacing at least one of a character and a keyword with an identifier.

Abstract: Provided is a method of data storage, the method including identifying a plurality of transactions in a pending queue, the transactions having one or more key value updates respectively corresponding to a plurality of keys, identifying a commonly associated key of the plurality of keys associated with commonly associated key value updates of the key value updates belonging to different ones of the transactions, respectively assigning transaction group IDs to the transactions based on respective transaction IDs assigned to the transaction group IDs, grouping the transactions into a respective transaction group of a plurality of transaction groups based on the assigned transaction group ID, and merging conflicting data writes corresponding to the commonly associated key value updates of the commonly associated key for grouped transactions of the transactions that are in a same one of the transaction groups.

Abstract: Methods and systems automatically adjusting resources and monitoring configurations of objects of a distributed computing system in response to changes to application programs. Methods search event messages for information indicating a change in execution of an object. The information is used to determine resource allocation rules of infrastructure resources by and a monitoring configuration for the object. Expected impacts on the infrastructure resource are determined from the rules. When an expected impact is greater than an associated impact threshold, use of the infrastructure resources may be adjusted to accommodate the changes. The adjustments include scaling up or down the infrastructure resources. When the object is a virtual object, the virtual object may be migrated from one server computer to another server computer within the distributed computer system. The monitoring configuration is used to adjust tools that monitor the objects of the distributed computing system.

Abstract: A computing system is provided, including a processor configured to receive a directed acyclic graph (DAG) template specifying a data pipeline of a plurality of processing stages. For each processing stage, the processor may be further configured to select a respective processing device of a plurality of communicatively linked processing devices. The processor may be further configured to determine a routing sequence between the plurality of processing devices according to the DAG template. The processor may be further configured to transmit a plurality of input packets encoding the plurality of processing stages to the respective processing devices selected for the processing stages as specified by the routing sequence. In response to transmitting the plurality of input packets, the processor may be further configured to receive, from a processing device of the plurality of processing devices, one or more output packets encoding a processing result of the data pipeline.

Abstract: The present approach relates to querying a configuration item (CI). The CI may be associated with a monitoring agent. In accordance with this approach, a check request is sent to a management, instrumentation, and discovery (MID) server. The check request may be sent to the monitoring agent. Additionally, a check result may be received from the monitoring agent. The check result may have data associated with the CI and the check request.

Abstract: An apparatus to facilitate data intelligent dispatching is disclosed. The apparatus includes one or more processing units including a plurality of execution units (EUs) to execute a plurality of processing threads and collection logic to collect statistics data for threads executed at the processing unit during execution of an application, and dispatch logic to dispatch the threads to be executed at a subset of the plurality of EUs during a subsequent execution of the application based on the statistics data.

Abstract: A computing system includes an application configured to request execution of at least one translation including at least one command. A first coupling facility is configured to perform a first modification process to modify a first structure based on a received command associated with an ongoing transaction. A second coupling facility includes a secondary circular queue loaded with first data blocks indicating the first modification process, and is configured to output a message response block (MRB). The application determines a most recent modification process performed by the secondary coupling facility based on the MRB.

Abstract: Examples of the present disclosure describe systems and methods for detecting and mitigating stack pivoting using stack artifact verification. In aspects, function hooks may be added to one or more functions. When a hooked function executes, artifacts relating to the hooked function may be left on the stack memory (“stack”). The location of the artifacts on the stack may be stored in a local storage area. Each time a hook in a hooked function is subsequently executed, protection may be executed to determine whether an artifact remains in the location stored in the local storage area. If the artifact is no longer in the same location, a stack pivot may be detected and one or more remedial actions may be automatically performed.

Abstract: The present technology pertains to an organization directory hosted by a synchronized content management system. The corporate directory can provide access to user accounts for all members of the organization to all content items in the organization directory on the respective file systems of the members' client devices. Members can reach any content item at the same path as other members relative to the organization directory root on their respective client device. In some embodiments novel access permissions are granted to maintain path consistency.

Abstract: Apparatuses, systems, and techniques to parallelize operations in one or more programs with data copies from global memory to shared memory in each of the one or more programs. In at least one embodiment, a program performs operations on shared data and then asynchronously copies shared data to shared memory, and continues performing additional operations in parallel while the shared data is copied to shared memory until an indicator provided by an application programming interface to facilitate parallel computing, such as CUDA, informs said program that shared data has been copied to shared memory.

Abstract: A method of avoiding throughput penalties imposed by SaaS vendors on a user group due to excessive API events from users in the group, monitoring API event rate or volume in time for requests from the group, collectively, and from individual users in the user group to a SaaS vendor is disclosed. Also, recognizing a power user as submitting API events in excess of a limit and taking action to reduce the user's impact on the API event rate of the group when the API rate for the group, overall, exceeds or approaches a SaaS imposed trigger of a throughput penalty on the group. Further included is rationing transmittal of API event submissions from the power user to the SaaS and avoiding triggering of the throughput penalty by the SaaS, reducing latency for the users in the group other than the power user and increasing latency for the power user.

Abstract: Embodiments may include techniques for hierarchical evaluation of the anomaly level of a system and its sub-components using domain knowledge, so as to provide improved accuracy and explainability compared to conventional methods. For example a method of anomaly detection in a hierarchical computer network may comprise defining a tree-like topological structure which describes how the hierarchical computer network comprises sub-components, wherein each node of the tree-like topological structure represents a sub-component of the hierarchical computer network, and wherein at least some of the sub-components are monitored to generate signals indicating an operational condition of each sub-component, collecting a plurality of time-series of maximum absolute anomaly scores for each monitored signal, and computing an anomaly score for a root node of the tree-like topological structure.