A data centres door survey that starts from the record the audit must see, then works back to the reader, the lock body and the frame fit.

Опубликовано: ноя 21, 2022

Smart Locks for Data Centres and Server Rooms: Door Survey and Fit Reality

Smart Locks for Data Centres and Server Rooms: Door Survey and Fit Reality

Data centres door survey is what this piece covers for this kind of building. A request arrives on a Tuesday morning and it is one line long: access records for the white-space entrance on a given date. The estate already carries electronic readers on that door. Nobody can answer the request, because nobody ever decided what the record was meant to prove — the credential presented at the reader, the trail the system kept afterwards, or the state of the door itself at that moment. Those are three different products, and each one implies a different door specification.

Direct answer. The one-line answer — data centres door survey is the working record for data centres door survey. data centres door survey is worthless until data-centres is attached to it as evidence. A data centres door survey names the door classes in one building. Each class needs its own specification. data centres door survey is the first one data centres door survey has to settle.

  • Split the door types — sort the door population into classes that behave alike, starting from data centres door survey
  • Match each type to a class — state what data-centres needs before anything is quoted
  • Evidence the fit — record the measurement that closes data centres door survey
  • In data centres door survey, this sequence is what data centres is checked against.

That is the argument of this guide. A data centres door survey is not a measurement trip with a lock catalogue in the van. It is the discipline of deciding what the record must prove, door by door, and then working backwards: from the record to the event, from the event to the credential, from the credential to the reader, from the reader to the panel, the lock body, the frame and the power that feeds them. The measurements come, but they come late, and only for the leaves the record decision has already selected.

This is the survey that works when server room access control has become part of the audited estate rather than a line in a fit-out budget. The reader is the visible part, and it is the least informative part. A property manager who starts at the reader will end up with a very tidy sheet of door dimensions and an estate that still cannot answer the request that started the whole exercise.

What follows is organised as a chain of decisions rather than a list of measurements. Each section asks one question and states what the survey has to produce before the next question can be answered.

What the Record Has to Prove Before a Reader Is Chosen

Surveyors in this vertical tend to start at the secure perimeter, because that is where the security expectation is loudest and where a visible reader settles a conversation. The audit trail requirement starts somewhere else: with the question the assurance reviewer will ask at the end of the year.

That question is narrow and it has three possible answers. Did a named person present this credential at this door? Did the system keep an event trail that can be reconstructed across every door in a zone? Did the door's own state at that time get recorded, meaning open, closed, forced or held open? Answer it and the rest of the survey writes itself. Leave it open and every later measurement is guesswork dressed up as rigour.

graph LR
  A[Audit question arrives] --> B[Record type required]
  B --> C[Event the door must emit]
  C --> D[Credential class that produces it]
  D --> E[Reader and lock body]
  E --> F[Leaf and frame fit]
  F --> G[Power and network path]
  G --> H[Evidence pack for the supplier]
  F --> I[Plant room and loading bay break the pattern]
  I --> J[Separate door class and separate evidence]

Read the diagram as the order of questions rather than the order of shopping. A reader chosen before the record decision inherits a specification nobody has justified: it captures the wrong events, and it sits on a frame that may have no room for the panel at all. The diagram also carries the warning that the estate is not one door class. The white space and the perimeter behave like a building; the plant room and the loading bay behave like a yard with a building inside it. The hardware behind this is in the hotel door locks.

One Door, From the Outside

Take the white space entrance door of an illustrative colocation hall. The survey starts on the approach side, because that is where the first decisions live: what the reader is, how high it sits, whether the controller and the network drop fall inside the secure line, and whether the badge is presented outside the glass so that someone following close behind cannot reach the reader through the door leaf.

The leaf is read next. Its thickness is captured at the latch edge and at the hinge edge, because a leaf that has been re-hung in a new frame is rarely square and the difference matters to a mortise case. The vision panel, if the leaf has one, decides whether the reader can be mounted on the frame at all or has to sit in a patch fitting on the glass. The closing device sets the swing and holds the leaf against the latch, which is where the survey learns whether an open door is a fault or a design condition.

From the inside face, the survey records what the corridor cannot show: whether there is a second reader or handle on the secure side, whether a mechanical override cylinder is present and where its key is held, and whether the escape arrangement at that door exists at all. That last item belongs to the fire strategy rather than to the lock, and it is written into the sheet as a question with a named owner rather than as an answer.

Written in that order, one door produces one line of record, and the line can travel. A supplier can answer it without walking the site, and a property manager can compare two doors in the same zone without standing in front of either of them.

Four Estate Zones, Four Record Obligations

The vertical facts for data centres describe four doors and four different kinds of door: a secure perimeter, a white-space entrance, a plant room and a loading bay. Treating them as one door with four finishes is the most common way a data centre door survey produces a specification that cannot be installed.

Estate zoneThe record it has to produceWhere the fit risk sitsThe team that owns the answer
Secure perimeterWhich credential was presented, and at which perimeter doorReader mounting on a hardened frame, the strike and the closing deviceSecurity operations
White-space entranceA trail across the white space for a named person, plus the door stateVision panel, rebate depth, whether a second reader is needed on the secure sideFacilities and security together
Plant roomEntry by exception, with a record that still reads correctly a year laterNarrow-profile steel leaf, plant vibration, no power or network close byMechanical or engineering
Loading bayContractor and delivery access inside a stated windowRoller or sectional leaf, weather exposure, vehicle impactLogistics with security

Two rows on that table will not accept the same hardware, and the reason is not aesthetic. The white-space entrance has to produce a per-person trail, which means a credential class, an event type and a controller that holds the trail. The loading bay has to produce a time window, which a roller door answers with a key handover and a supervisor's log rather than with an event trail. Surveying them with the same sheet produces a specification that satisfies neither.

The Credential Owners and the Record They Leave Behind

A badge with zone rights is an organisational fact wearing a plastic coat. Before any door is walked, the buyer needs to know who issues the credential, who holds the highest level of authority, how a zone right is withdrawn when somebody leaves, and where the master credential physically lives on a Tuesday night.

The audit trail for access records is only as useful as that organisation allows it to be. Zone rights decide what the trail can actually say about a person, which is why they belong in the survey sheet rather than in a separate access policy that arrives after the hardware has been chosen.

A survey line that says badge reader at the perimeter has recorded nothing about zone rights. Zone rights are what make a perimeter door mean anything at all: without them the reader is an expensive doorbell. The honest survey output is a statement of which zone a door opens, which credential classes may be presented there, and who in the building is accountable for that list.

Records about named people are personal data, and their retention, export and readership are decisions that sit above the lock. The survey records the question and the owner; it does not settle the answer.

CONFIRM-WITH-LOCAL-DATA-PROTECTION-REVIEWER: retention periods, export rights and who may read the access trail for a named person are decided by the data-protection reviewer for the estate, not by the lock supplier.

Two related questions are worth reading before the survey closes, because they shape which events the reader has to emit in the first place. The record question is set out in Audit Trail Requirements for Data Centres and Server Rooms: What Records Matter, and the retention question is handled in Data Retention and Privacy for Smart Locks in Data Centres and Server Rooms.

Where the Leaf Ends and the Frame Begins

This is the part of the survey that attracts the most attention and deserves the least confidence on its own. Door fit measurement is straightforward technique and it is well documented: Smart Lock Door Measurement Guide covers the sequence, and the individual measurements have their own references in Smart Lock Door Thickness Measurement, Smart Lock Backset Measurement, Smart Lock Mortise Measurement, Smart Lock Door Gap Measurement and Smart Lock Strike Plate Measurement.

What the technique does not tell you is which range a measurement has to fall inside for a particular model. The door fit field is populated only for some models in the catalogue, and the ranges differ by family: models built for indoor wooden doors record a narrower thickness band than the families built for hotel and metal doors, one narrow-profile model records a lock case of 35 by 85 mm or 40 by 85 mm, and one fingerprint mortise model records a faceplate of 227 by 66 by 20 mm. Treat those as the shape of the problem rather than as an answer for the estate, and confirm on the dated product page for the specific SKU before a measurement is allowed to become a specification.

Two data centre realities sit on top of the technique. A retrofit estate often contains leaves that are not the ones the frame was built around, so the survey marks every leaf as original or replaced and treats a replaced leaf as voiding the frame assumption. And the frames on plant room and loading bay doors are frequently surface-mounted or hollow, so the survey reads the frame rebate and the strike fixing rather than assuming a standard timber rebate exists behind the door.

Readers, Panels and the Geometry That Fights Back

The reader is a hardware object that has to live on a surface, and the surfaces in this vertical are unforgiving. Secure perimeter door hardware has to sit on powder-coated steel, brushed stainless or glass, and the panel needs a flat, solid, uncoated area with the cable route behind it and the lock body on the other side of the leaf.

The object list is longer than most first surveys allow for. A reader, an escutcheon or panel, a lever or handle, a cylinder and rose, a strike plate and keeper, a mortise case, a spindle, a motor and gearbox, a radio module, a network gateway and a hold-open device all have to be accounted for before a door can be called specified.

The white space adds a class of door that the survey must name explicitly: the equipment cage and its man-door, and the cabinets inside it. These are not room doors and they are not cabinet lock bodies either. The compatibility notes for Door Types in Data Centres and Server Rooms: Cages, Cabinets and Man-Doors Compatibility Notes set out how those three behave against a lock body, and they are worth reading before the survey sheet is drawn.

Power, Cabling and the Doors Nobody Wants to Touch

The plant room is where a data centre survey goes wrong by omission. Plant room door hardware has to be chosen on the assumption that nobody is standing next to it. The room sits on the far side of the secure line, often a long walk from any desk, sometimes with no network drop and no charger, and it holds the kind of equipment that makes its own environmental demands. The reader there will be battery powered in most cases: several catalogue families record four AA cells and a low-battery alert, and some record an emergency power override through a USB terminal.

A survey that records battery without recording where the change happens has recorded a cost rather than a plan. The sheet needs a line for the route, a line for who carries the cells, and a line for what happens to the door record when the cells run down at three in the morning. The programme side of that question belongs to the Battery and Power Programme for Smart Locks in Data Centres and Server Rooms.

The loading bay is the other door nobody wants to touch, because it takes weather, vehicles and deliveries. A roller or sectional leaf is not a door leaf with a lock case in it, and the honest survey output for that opening is a statement that the leaf type does not accept the hardware, together with what the site does instead. The abuse and load cases for these openings are set out in Vandalism, Misuse and Environmental Load on Locks in Data Centres and Server Rooms, and the survey should not try to duplicate that argument — it should record the exposure and point at it.

The Decisions That Belong to the Authority, Not to the Catalogue

Some of what a survey encounters cannot be answered by any supplier, however good the catalogue is. The escape arrangement on a perimeter door, the panic hardware on the secure side of a white-space entrance, the swing of a leaf on an escape route and the use of a hold-open device are all decided by the fire strategy for the building and the local authority that approves it.

CONFIRM-WITH-LOCAL-FIRE-AUTHORITY: escape routes, panic and exit hardware, hold-open devices and any locking arrangement on an escape route are decided by the local fire authority and the building fire strategy, not by the lock supplier.

CONFIRM-WITH-LOCAL-BUILDING-AUTHORITY: frame construction, door width and any alteration to a rated assembly are decided by the local building authority and the fire strategy for the building.

The survey's job in this area is narrow but important: it records the question, names the door it applies to, and names the authority or internal owner who must answer. A survey sheet that returns with a tick in a box and no name attached is not evidence, and it is the easiest defect in the whole pack to fix while the crew is still on site.

The Evidence Pack a Supplier Can Price

A survey earns its cost when it can leave the site. The pack that travels is a door list, one line per leaf, and it has to be answerable by someone who will never see the building. That means the sheet asks the supplier five questions, and each one has to be answered with a dated document for the exact model rather than with a family brochure.

Question the buyer asksEvidence to demandWhat it settles before an order
Will this reader produce the event the audit needsThe event list for that model on its dated product pageWhether the record decision survives contact with the product
Does the lock body fit the leaf and the frameDimension drawings for the specific SKUWhether the door can be retrofitted at all
What powers it and how is it changedThe power, low-battery and emergency power fields for the specific SKUWhere the maintenance route has to go
How is it mounted and what body does it carryThe panel, body and mortise records for the same modelWhether two doors in one zone can differ
What happens when the system or the network is downThe override and standalone behaviour for that modelWhether a manual route exists and who holds it

The catalogue itself offers the documentary route most buyers need: dimension drawings, an installation guide and a full datasheet are available on request for each model, with logo, finish, packaging, manuals and product configuration handled for project orders. The buyer's job is to require those documents dated to the model being quoted, so that two suppliers answer the same door line with comparable evidence.

That evidence then becomes the RFQ. The template in RFQ Template for Smart Locks for Data Centres and Server Rooms is built to receive a completed door list, and the sharper question set behind it sits in Questions to Ask a Supplier Before Choosing Locks for Data Centres and Server Rooms. Neither is much use without the survey sheet behind it.

Three Surveys That Went Wrong

The first failure is the one the vertical facts warn about directly: a reader was bought before anyone decided what the record had to prove. The result is a beautiful door that captures events nobody will ever query, and a trail that cannot answer a request.

The second failure is one specification for the whole estate. A perimeter frame, a white-space leaf and a plant room door were surveyed as a single hardware family because the survey walked the building in one direction and stopped writing after the first zone. The order of the walk decides the order of the specification.

The third failure is quieter. The survey was run by the team that owns the building, and the people who issue credentials, hold the master and answer for the zone hierarchy were not in the room. The door list came back complete on fit and empty on access, because the questions about who may open which door cannot be answered by a tape measure. The remedy is procedural rather than technical: the credential owners sit down with the sheet before the walk, not after it.

Where the Data Centre Survey Hands

A completed data centres door survey is the input to three larger decisions, and it is worth naming them so the survey does not try to settle them. Master key and credential control is a separate exercise, set out in Smart Lock Key Control and Master Key Decisions for Data Centres and Server Rooms. Sequencing a multi-site estate is set out in Phased Retrofit: Smart Locks Across a Data Centres and Server Rooms Portfolio. Choosing between suppliers on criteria rather than on a unit price is set out in Smart Lock Vendor Selection Criteria for Data Centres and Server Rooms.

What the survey itself owns is narrow and durable: which door classes exist, what record each one has to produce, what the fit reality is at each leaf and frame, and who must answer the questions that no supplier can answer. Written that way, the sheet survives a change of building, a change of access platform and a change of the people who hold the credentials.

When the door list is ready, it is worth sending it as a document rather than as a summary of a site visit. A supplier working from the door classes, the record requirements and the per-leaf measurements can return fit options and documentation against each model, and the property manager can compare doors that were never opened on the same morning. The request a quote page is the place to send it, with the zone names and the record requirement for each door written into the body of the message.

The estate that settles the record question first will spend far less of its time arguing about readers afterwards.

Ваша корзина
Ваш опыт использования этого сайта будет улучшен, если вы разрешите использование файлов cookie Политика использования файлов cookie
Chat on WhatsApp